Phosphotyrosine-Dependent Targeting of Mitogen-Activated Protein Kinase in Differentiated Contractile Vascular Cells
Abstract Tyrosine phosphorylation has been linked to plasmalemmal targeting of src homology-2–containing proteins, activation of mitogen-activated protein (MAP) kinase, nuclear signaling, and proliferation of cultured cells. Significant tyrosine phosphorylation and MAP kinase activities have also been reported in differentiated cells, but the signaling role of tyrosine-phosphorylated MAP kinase in these cells is unclear. The spatial and temporal relation between phosphotyrosine and MAP kinase immunoreactivity was quantified in differentiated contractile vascular smooth muscle cells by using digital imaging microscopy. An initial association of MAP kinase with the plasmalemma required upstream protein kinase C activity but occurred in a tyrosine phosphorylation–independent manner. Subsequent to membrane association, a delayed redistribution of MAP kinase, colocalizing with the actin-binding protein caldesmon, occurred in a tyrosine phosphorylation–dependent manner. The apparent association of MAP kinase with the contractile proteins coincided with contractile activation. Thus, tyrosine phosphorylation appears to target MAP kinase to cytoskeletal proteins in contractile vascular cells. This targeting mechanism may determine the specific destination and thereby the specialized function of MAP kinase in other phenotypes.
- signal transduction
- vascular smooth muscle
- tyrosine phosphorylation
- mitogen-activated protein kinase
Stimulation of cultured cells with mitogens increases the tyrosine kinase activity of plasmalemmal growth factor receptors.1 Phosphorylation of tyrosine residues serves to target intracellular protein kinases that contain src homology-2 (SH2) domains to the surface membrane, where they initiate a cascade of events leading to cell proliferation.2 3 Mitogen-activated protein (MAP) kinase has been recognized as one of the major targets for tyrosine phosphorylation that plays a pivotal role in the transduction of extracellular mitogenic signals to the nucleus. MAP kinase is a proline-directed Ser/Thr protein kinase, which is fully activated by dual phosphorylation at Thr and Tyr residues.3 Activated MAP kinase has been shown to translocate into the nucleus,4 activate nuclear transcription factors, and stimulate growth of undifferentiated cells.1 2 3 4 5 Significant protein tyrosine kinase and MAP kinase activities have recently been reported in adult, terminally differentiated T lymphocytes,6 human platelets,7 neuronal cells,8 and smooth muscle cells9 10 11 upon stimulation with nonmitogenic agonists. However, the intracellular location of tyrosine phosphorylation and MAP kinase as related to their signaling role to invoke specialized functions in nonproliferative cells, in general, and contractile cells, in particular, remains unclear.
In ferret aortic cells, a significant component of the contraction to phenylephrine persists in Ca2+-free bathing solution and also in the absence of increases in [Ca2+]i.12 This Ca2+-independent contraction is dependent on protein kinase C (PKC) activity, presumably ε-PKC,13 and has been associated with MAP kinase stimulation.14 However, the degree to which tyrosine phosphorylation of MAP kinase plays a signaling role in this Ca2+-independent contraction has not been determined. In the present study, we used digital imaging analysis to determine tyrosine phosphorylation and MAP kinase trafficking during contractile activation of intact vascular smooth muscle cells.
Materials and Methods
Ferrets were anesthetized with chloroform and dissected following procedures approved by the Institutional Care and Use Committee. The thoracic aorta was removed to a dissection dish filled with oxygenated Krebs’ solution. The aorta was cleaned of connective tissue, and the endothelium was removed by gentle rubbing.
Tissue samples were frozen with Freon, precooled in liquid N2, and stored at −80°C until used. Samples for phosphotyrosine (P-Tyr) immunoblots were thawed in an acetone–dry ice slurry and then homogenized in buffer containing 50 mmol/L Tris (pH 7.4), 10% glycerol, 5 mmol/L EGTA, 140 mmol/L NaCl, 1% Nonidet P-40 and the protease inhibitors leupeptin and pepstatin (5.5 μmol/L, Boehringer Mannheim), aprotinin (20 kallikrein inhibitory units [KIU], Sigma Chemical Co), 1 mmol/L Na3VO4, 10 mmol/L NaF, 0.25% (wt/vol) sodium deoxycholate, 100 μmol/L ZnCl2, 20 mmol/L β-glycerophosphate, and 1 mmol/L phenylmethylsulfonyl fluoride (PMSF). Before immunoblotting, the P-Tyr samples were immunoprecipitated with a MAP kinase antibody covalently bound to protein A–agarose beads (UBI). Other samples for immunoblots were homogenized in buffer containing 20 mmol/L MOPS, 10% glycerol, 2% SDS, 5 mmol/L EGTA, 5.5 μmol/L leupeptin and pepstatin, 20 KIU aprotinin, 20 μmol/L PMSF, 100 μmol/L ZnCl2, and 10 mmol/L dithiothreitol. Protein-matched samples were subjected to electrophoresis on 8% SDS-polyacrylamide gels and then transferred electrophoretically to nitrocellulose membranes. The membranes were incubated in 5% dried milk in PBS-Tween buffer at 22°C for 1 hour, washed with PBS-Tween three times for 5 minutes each, and then incubated in the presence or absence of specific primary antibody solution at 4°C overnight. The membranes were washed five times for 15 minutes each and then incubated in horseradish peroxidase–conjugated IgG (Bio-Rad) for 1.5 hours. The blots were washed with PBS-Tween five times for 15 minutes each and visualized with an Enhanced Chemi-Luminescence detection system (Amersham). PBS-Tween contained (mmol/L) Na2HPO4 80, NaH2PO4 20, and NaCl 100, along with 0.05% Tween.
Single ferret aortic cells were enzymatically isolated using a procedure developed specifically to retain their pharmacological responsiveness.15 Briefly, aortic strips (50 mg) were placed in a digestion medium containing 4.0 mg elastase (3.3 U/mg, Boehringer Mannheim), 4.4 mg CLS 2 collagenase (171 U/mg, Worthington), and 3.6×10−2 mol/L trypsin inhibitor (type II-S soybean, Sigma) and dissolved in 7.5 mL Ca2+- and Mg2+-free Hanks’ solution. The preparation was incubated in a shaking water bath at 34°C in an atmosphere of 95% O2/5% CO2 for 100 minutes. The quality of freshly enzymatically isolated mammalian smooth muscle cells is inversely proportional to the yield of the cells. Hence, the very gentle isolation method we use produces only few cells per coverslip. Centrifugation and aspiration of the tissue through wide-bore pipettes (methods commonly used to increase cell yield) are not used, because these procedures have been found to cause irreversible damage to the cells. Rather, cells are concentrated by letting them settle by gravity onto glass coverslips, since only healthy cells adhere to glass.15
Cells were incubated in Ca2+-free (2 mmol/L EGTA) Hanks’ solution for 15 minutes and then treated with phenylephrine (10−5 mol/L) for the time indicated. Cells were fixed in 2% paraformaldehyde, permeabilized with 0.1% Triton X-100, blocked with 2% goat serum, and then reacted with a polyclonal anti–MAP kinase antibody raised in rabbit (UBI, 1:1000), followed by fluorescein isothiocyanate (FITC)–conjugated anti-rabbit IgG (Sigma, 1:100). Fluorescent images were viewed through a Nikon ×100 oil-immersion objective (numerical aperture, 1.3) on a Zeiss IM35 microscope; a fluorescein filter set (excitation, 485±5.2 nm; dichroic, 500 nm; emission, 530 nm) was used. The ×100 oil-immersion objective (numerical aperture, 1.3) used to obtain high-resolution details of structures of the cell and to perform quantitative digital imaging analysis allowed the recording of only one cell image at a time. All cells on the coverslip did fluoresce, as determined by scanning with a lower-power objective, and did show qualitatively similar patterns although with less detail. For colocalization studies of MAP kinase and P-Tyr, the same cells were reacted with monoclonal anti–P-Tyr antibody raised in mouse (Sigma, 1:500), followed by Texas Red–conjugated anti-mouse IgG (Calbiochem, 1:100). Fluorescent images of P-Tyr were viewed by using a Texas Red filter set (excitation, 560±20 nm; dichroic, 595 nm; emission, 635±30 nm). For colocalization studies of MAP kinase and caldesmon, the cells were reacted with a monoclonal rhodamine-labeled anti–gizzard-caldesmon antibody (CD5, 1:500) raised in mouse. Rhodamine-conjugated caldesmon antibody (CD-5) was prepared as described by Khanna and Ullman.16 The unbound dye was removed by G-50 Sephadex column chromatography, followed by dialysis. Complete removal of free rhodamine in the preparation was ascertained by the absence of an absorption peak at 518 nm. Fluorescent images of caldesmon were viewed by using a rhodamine filter set (excitation, 546±12 nm; dichroic, 580 nm; emission, 590 nm). The concentrations of the different antibodies were titrated in order to obtain roughly comparable fluorescence intensities. Under the same conditions used for colabeling, cells labeled with FITC alone have shown no detectable signal with either the rhodamine or Texas Red filter sets. Similarly, when cells were labeled either with rhodamine-conjugated anti-caldesmon antibody alone or with the anti–P-Tyr primary antibody and the Texas Red–labeled secondary antibody, the cells did not show any detectable signal when a fluorescein filter set was used.
Digital Imaging Microscopy
Images were acquired with a liquid-cooled digital CCD camera (model CH250, Photometrics) and digitized by using pmis image processing software (Photometrics) attached to an MS-DOS–based microcomputer. All images were acquired identically by using the same camera gain and integration time. The digitized images were transferred to a Sun Spark IPC computer for analysis. Images were corrected for dark current, flat-field–corrected, and background-subtracted. Fluorescent beads (140 nmol/L, Molecular Probes) were added to the mounting solution and imaged identically to determine the point-spread function of the microscope. For each cell, 11 optical sections at 0.4-μm intervals were acquired. The image of each cell section was deconvolved by using an image restoration algorithm based on the nearest neighbor theory and a quadratic regularization function.17 The noise variance was calculated for each individual cell image, whereas the regularization parameter (λ) was set constant at 2×10−4 for all images, which maximized the pixel-to-pixel contrast necessary for colocalization calculations.
The average cell fluorescence was measured by integrating the total cell fluorescence intensity and dividing by the number of pixels in the cell image. Where surface-to-cytosol fluorescence ratios were calculated, a peripheral area 1/10 of the cell thickness was defined, and the average surface fluorescence was calculated by dividing the integrated pixel intensity of the peripheral area by the number of pixels in the area. We have previously performed studies with the plasma membrane marker 7-decylBODIPY-1-propionic acid to confirm our ability to define the cell periphery.18 The average cytosolic signal was calculated for the remainder of the cell image. The average surface membrane signal and the average cytosolic signal were used to calculate the surface-to-cytosol fluorescence ratio. By using the surface-to-cytosol fluorescence ratio for comparison, the effects of the difference in cell thickness and fluorescence labeling in different cells were minimized.
In the colocalization studies, the observed overlap was defined as the percentage of pixels containing the first protein that also contained the second protein and was determined by using a logical AND of the two images. The predicted overlap due to chance was calculated by generating, using a gaussian distribution, random images containing the same number of positive pixels as in the observed images.
The changes in cell length were measured in activated live cells observed under the microscope with a Nikon ×20 objective. The magnitude of cell shortening was expressed as the final cell length as a fraction of the initial cell length.
To prevent excessive cell shortening from obscuring the fluorescent images, all cells intended for imaging were pretreated for 5 minutes with hypertonic Hanks’ solution containing 330 mmol/L sucrose. We have previously demonstrated that identical kinetics and magnitude of PKC translocation are obtained when staurosporine (an inhibitor of the catalytic activity of the protein kinase) is used to inhibit cell shortening or when the early time points preceding cell shortening are compared in the absence of any inhibitor.13 14 18 19 20 The hypertonic solution is thought to prevent contraction at the crossbridge level21 22 and has also been shown previously in ferret aortic cells not to interfere with depolarization and agonist-induced increases in [Ca2+]i20 or with the distribution of actin and the actin-binding protein calponin in ferret portal vein cells.23 Both the control resting cells and the stimulated cells were treated identically with the hypertonic solution.
Data were analyzed and presented as mean±SEM unless indicated otherwise. Data were compared by using Student’s t test for unpaired data.
Krebs’ solution contained (mmol/L) NaCl 120, KCl 5.9, CaCl2 2.5, MgCl2 1.2, NaHCO3 25, NaH2PO4 1.2, and dextrose 11.5 at pH 7.4 when bubbled with 95% O2/5% CO2. Hanks’ solution contained (mmol/L) NaCl 137, KCl 5.4, dextrose 5.55, NaHCO3 4.17, Na2HPO4 0.42, KH2PO4 0.44, MgCl2 1.4, and CaCl2 1.0 at pH 7.4. In Ca2+-free Hanks’ solution, CaCl2 was replaced by 2 mmol/L EGTA. High-K+ solution was prepared by equimolar substitution of NaCl with KCl.
l-Phenylephrine HCl was purchased from Sigma; staurosporine and calphostin C, from Kamiya; and tyrphostin, from GIBCO BRL. RG 50864-2 was a generous gift from Rhone-Poulenc Rorer, and ST638 was generously provided by Kaneka.
Western blot analysis (Fig 1⇓) indicated that the anti–MAP kinase antibody was reactive to both a 44- and a 42-kD species. Characteristically, however, the lower band was much fainter than the upper band. These results are consistent with the results of Childs et al11 in differentiated smooth muscle from the rat aorta, where the 44-kD isoform of MAP kinase was found to be more enriched than the 42-kD isoform. The anti–P-Tyr antibody gave detectable staining in only one band at ≈44 kD. The anti-caldesmon antibody gave two bands at ≈130 and 80 kD. Whether the lower band indicated the presence of the nonmuscle isoform of caldesmon or a degradation product of the larger isoform was not further investigated. No detectable bands were observed in the absence of the primary antibodies.
In resting ferret aortic cells, a significant MAP kinase–specific fluorescence, homogeneously distributed in the cytosol but excluding a central region corresponding to the nucleus, was seen (Fig 2⇓). In colabeled cells, parallel quantification of total cellular P-Tyr–specific fluorescence showed little or no signal (17.8±1.6 units) in unstimulated cells. Phenylephrine caused a significant increase in the total cellular P-Tyr signal (Fig 2⇓), which reached a maximum of 163.2±9.7 units at 4 minutes and a plateau of 122.7±20.8 units by 10 minutes (Fig 3A⇓, solid circles). In cells stimulated with phenylephrine for 4 minutes, both MAP kinase and P-Tyr colocalized near the cell membrane. After 10 minutes of stimulation, P-Tyr and MAP kinase colocalized to filamentous structures along the longitudinal axis of the cell (Fig 2⇓).
Control experiments were performed, after 10 minutes of exposure of the cells to phenylephrine, in the absence of primary antibody, and competitive studies were performed in the presence of primary antibody and either competing P-Tyr or a competing peptide. For P-Tyr the average fluorescence intensity was 122.7±20.8 units (n=9) in the presence of the primary and secondary antibodies, 24.43±2.27 units (n=7) in the presence of the primary and secondary antibodies and 1 μmol/L O-phospho-dl-tyrosine (Sigma), and 3.15±0.03 units (n=7) in the presence of secondary antibody alone. For MAP kinase the average fluorescence intensity was 172.36±9.42 units (n=22) in the presence of the primary and secondary antibodies, 32.63±2.67 units (n=8) in the presence of the primary and secondary antibodies and a competing peptide to which the primary antibody was raised, and 6.67±1.70 units (n=7) in the presence of the secondary antibody alone. We have performed additional control experiments to rule out nonspecific cross-reactivity between primary and secondary antibodies, in which the cells were treated with monoclonal anti-caldesmon antibody followed by FITC-labeled goat anti-rabbit IgG. We have also performed experiments using polyclonal anti–MAP kinase antibody raised in rabbit, followed by sheep anti-mouse IgG. No significant staining was detectable under these conditions.
The kinetics of the phenylephrine-induced redistribution of P-Tyr and MAP kinase immunofluorescence were compared in different cells by quantifying the ratio between the surface membrane fluorescence and the cytosolic fluorescence at different time points of activation. Phenylephrine caused a transient peak at 4 minutes in localization of P-Tyr signal at the surface membrane, but by 10 minutes the surface-to-cytosolic ratio returned to basal levels (Fig 3B⇑, solid circles). Likewise, the MAP kinase signal maximally translocated to the cell membrane at 4 minutes of stimulation, but the ratio declined to basal levels at 10 minutes (Fig 3C⇑, solid circles).
We tested whether the similarities in the spatial distribution and kinetics of P-Tyr and MAP kinase immunofluorescence are due to actual colocalization of P-Tyr groups and MAP kinase or simply due to chance. As illustrated in the Table⇓, the observed overlap between P-Tyr and MAP kinase in unstimulated cells was not significantly different from the predicted overlap due to chance. However, a significant increase in the observed overlap was demonstrated at both 4 and 10 minutes of stimulation.
We have previously shown, in the same cell type, that phenylephrine also causes activation and translocation of ε-PKC from the cytosol to the surface membrane.13 To test whether PKC activation is required for tyrosine phosphorylation and MAP kinase translocation, we investigated the effect of two chemically unrelated PKC inhibitors. Calphostin C competes at the diacylglycerol-binding site on the regulatory domain,24 whereas staurosporine competes at the ATP-binding site on the catalytic domain25 of PKC. As shown in Fig 3A⇑ (triangles), the phenylephrine-induced increase in the P-Tyr signal was significantly inhibited by both calphostin C and staurosporine. Also, the redistribution of the P-Tyr (Fig 3B⇑, triangles) and MAP kinase (Fig 3C⇑, triangles) signals was completely abolished by both PKC inhibitors.
To investigate whether tyrosine phosphorylation is necessary for MAP kinase translocation, we tested the effect of tyrphostin, a relatively specific tyrosine kinase inhibitor that competes with the phosphoryl acceptor substrate tyrosine.26 Tyrphostin pretreatment significantly inhibited the phenylephrine-induced increase in the P-Tyr signal (Fig 3A⇑, open circles), and no significant localization of P-Tyr at the surface membrane was observed (Fig 3B⇑, open circles). Surprisingly, in the presence of tyrphostin to inhibit tyrosine kinase activity, MAP kinase translocation to the surface membrane was actually augmented (Fig 3C⇑, open circles). In contrast, tyrphostin completely inhibited the delayed targeting of MAP kinase to the cytoskeleton (Fig 3C⇑, 4 to 10 minutes). Tyrosine kinase inhibitors from other sources, RG 50864-2 (Rhone-Poulenc Rorer)27 and ST638 (Kaneka),28 produced similar results.
We investigated the significance of P-Tyr–dependent signaling on the phenylephrine-induced Ca2+-independent contraction, which in control cells reached a maximal value by 10 minutes. As shown in Fig 3D⇑ (open circles), the Ca2+-independent contraction was significantly inhibited by the tyrosine kinase inhibitor tyrphostin (100 μmol/L). To avoid possible nonspecific effects not related to inhibition of tyrosine kinase, greater concentrations of tyrphostin were not used to inhibit cell contraction. We have used more potent tyrosine kinase inhibitors from other sources and have found that compounds such as RG 50864-2 and ST638 at 100 μmol/L completely abolished cell contraction. In contrast, tyrphostin caused no statistically significant inhibition of Ca2+-dependent contractions induced by membrane depolarization with high-K+ (66 mmol/L) solution (data not shown). We also found that the Ca2+-independent contraction was completely inhibited by the PKC inhibitors calphostin C and staurosporine (Fig 3D⇑, triangles).
The delayed phenylephrine-induced filamentous distribution of MAP kinase suggested its interaction with the contractile myofilaments. We imaged the distribution of MAP kinase and the actin-binding protein caldesmon and tested for their colocalization during activation by phenylephrine. As shown in Fig 4⇓, caldesmon is associated with a cytoskeletal structure, presumably actin filaments, along the longitudinal axis of the cell. We have compared the distribution of caldesmon and actin filaments and have found that both proteins show indistinguishable filamentous distribution along the longitudinal axis of the cell (Fig 4⇓, bottom). However, to determine exactly what protein on the filament is involved, detailed immunoelectron microscopy studies would be required. Phenylephrine did not cause any detectable qualitative changes in the distribution pattern of caldesmon. In unstimulated cells, ≈40% of MAP kinase colocalized with caldesmon, but this observed overlap was not significantly different from the predicted overlap due to chance (Table⇑). After 4 minutes of stimulation, the translocation of the MAP kinase signal to the surface membrane caused a significant drop in its observed overlap with the caldesmon signal (23.2%). After 10 minutes of activation, we found an increase in the observed overlap between MAP kinase and caldesmon signals that was significantly different from the predicted overlap due to chance (Table⇑).
The present study demonstrates tyrosine phosphorylation to be an important signaling event during Ca2+-independent activation of ferret aortic cells as shown by the following: (1) Phenylephrine caused a significant and sustained increase in P-Tyr immunoreactivity. (2) The time of maximal localization of P-Tyr to the myofilaments corresponded with the time of maximal phenylephrine contraction. (3) Tyrosine kinase inhibitors significantly inhibited the phenylephrine-induced increase in tyrosine phosphorylation. (4) Tyrosine kinase inhibitors significantly inhibited the Ca2+-independent phenylephrine-induced contraction but not the Ca2+-dependent contraction induced by membrane depolarization. These results are consistent with reports that the activity of pp60c-src tyrosine kinase is significantly elevated in ATP-stimulated aortic vascular smooth muscle9 and that tyrosine kinase inhibitors suppress agonist-mediated contraction but not Ca2+-mediated contraction in β-escin–permeabilized mesenteric microvessels.29 Taken together, these results support the hypothesis that tyrosine phosphorylation is a major signaling event during contractile activation of vascular smooth muscle cells.
An important finding of the present study is that MAP kinase initially translocates to the surface membrane but then undergoes a second redistribution to the cytoskeleton. The present results suggest that the two phases of MAP kinase redistribution are differentially regulated. The initial translocation of MAP kinase to the surface membrane appears to be dependent on PKC activation, since it was completely inhibited by PKC antagonists and has previously been shown to coincide with translocation of ε-PKC to the plasma membrane.14 Also, the observation that PKC antagonists significantly inhibited both tyrosine phosphorylation and MAP kinase redistribution suggests that PKC is acting upstream from these signaling events in the protein kinase cascade.
In contrast, the delayed localization of MAP kinase to the cytoskeleton appears to be regulated by tyrosine phosphorylation. This is supported by two observations. First, the similarities in the spatial distribution and kinetics of translocation of the P-Tyr and MAP kinase signals as well as their significant colocalization are consistent with the interpretation that the observed increase in tyrosine phosphorylation is coupled to MAP kinase. We cannot rule out the coexistence of additional tyrosine-phosphorylated proteins in the cells, but those proteins appear to be associated with MAP kinase. Similarly, we cannot rule out the possibility that the apparent cycling of MAP kinase might represent bulk cycling of membranes. Second, tyrosine kinase antagonists completely inhibited the delayed localization of MAP kinase to the cytoskeleton but not the initial plasmalemmal localization. In fact, in the presence of tyrphostin, MAP kinase translocation to the surface membrane was actually augmented, perhaps reflecting an increased lipophilicity of the nonphosphorylated kinase. The fact that tyrphostin treatment abolishes the second redistribution is consistent with a P-Tyr–dependent targeting mechanism of MAP kinase to the cytoskeleton.
It is well recognized that in proliferating cells, activation of growth factor receptors stimulates their intrinsic tyrosine kinase activity and subsequent autophosphorylation on tyrosine residues. Phosphorylated tyrosine residues serve as high-affinity binding sites for cellular proteins that carry SH2 domains.30 31 These SH2-phosphopeptide interactions recruit other signaling kinases to the receptor, where they can be phosphorylated. Thus, tyrosine phosphorylation has been thought to act as a targeting mechanism for translocation of kinase complexes to the surface membrane. Our results are not consistent with this picture, however, in that in these differentiated cells the plasmalemmal translocation of MAP kinase is only transient and tyrosine phosphorylation actually leads to redistribution of MAP kinase away from the surface membrane complex and toward the cytoskeleton.
The translocation of MAP kinase to the myofilaments does not appear to be a uniform response to all modes of stimulation in ferret aortic cells. We have previously shown that stimulation of the same cell type with a 66 mmol/L K+ depolarizing solution, a stimulus that causes a Ca2+-dependent contraction, does not cause any significant changes in the distribution of MAP kinase.14 We cannot rule out the possibility that other agonists that activate tyrosine phosphorylation in a Ca2+-dependent manner do not also cause targeting of MAP kinase to the cytoskeleton in differentiated smooth muscle cells. The main distinction we wish to make, however, is not between agonists but rather between differentiated (contractile) and undifferentiated (cultured) smooth muscle cells. Using the same batch of MAP kinase antibody, we have looked at the distribution of MAP kinase in undifferentiated vascular smooth muscle cells (R.A. Khalil, K.C. Kent, and K.G. Morgan, unpublished data, 1995). We have found (as have many other laboratories4 32 33 ) that potent mitogens, such as platelet-derived growth factor, basic fibroblast growth factor, and epidermal growth factor, cause significant translocation of MAP kinase into the nucleus. Nonmitogenic or less mitogenic agonists, such as phenylephrine, angiotensin II, and thrombin, also cause a slight increase in nuclear MAP kinase. With all the tested agonists, no localization of MAP kinase to the microfilaments was observed in undifferentiated vascular smooth muscle cells.
The observations that the phenylephrine-induced Ca2+-independent contraction of ferret aortic cells was completely abolished by PKC inhibitors and significantly inhibited by tyrosine kinase inhibitors suggest that PKC activation and tyrosine phosphorylation of MAP kinase are involved in a signaling cascade terminating in a Ca2+-independent smooth muscle contraction. Although activated MAP kinase can phosphorylate many proteins in vitro, the question arises as to the relevant intracellular protein substrate that leads to smooth muscle contraction. Caldesmon is an actin-binding protein34 35 that has been implicated in mediating a Ca2+-independent contraction of smooth muscle36 but has been reported to be present only in the subpopulation of actin filaments associated with myosin filaments.37 The present study showed, during steady state activation, an increase in the observed overlap between MAP kinase and caldesmon signals that was significantly different from the predicted overlap due to chance. These results are consistent with reports that caldesmon is phosphorylated in vivo by proline-directed kinases such as MAP kinase,10 11 38 a process that reverses the caldesmon-induced inhibition of actin-activated myosin ATPase activity34 and thereby would cause contraction of the smooth muscle.
Thus, in contrast to undifferentiated cells where tyrosine phosphorylation targets protein kinases toward the surface membrane, P-Tyr–dependent activation of MAP kinase in differentiated smooth muscle cells appears to target the kinase to the cytoskeleton. On the other hand, the trafficking of intracellular kinases outlined for proliferating cells cannot be applied to differentiated contractile cells, and it is possible that P-Tyr–dependent targeting of MAP kinase can perform additional specialized functions in other differentiated cell types.
This study was supported by Public Health Service grants HL-31704 and HL-42293 to Dr Morgan and AR-41637 to Dr Wang and by an American Heart Association fellowship to Dr Khalil. The authors thank Dr A. Zilberstein (Rhone-Poulenc Rorer, Collegeville, Pa) for providing RG 50864-2, Dr T. Shiraishi (Kaneka, Takasago Hyogo, Japan) for providing ST638, D. Fischer for typing and editing the manuscript, and R. Littlefield for expert photographic assistance.
This article was sent to Harold C. Strauss, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
- Received October 3, 1994.
- Accepted March 6, 1995.
- © 1995 American Heart Association, Inc.
Rossomando AJ, Payne DM, Webber MJ, Sturgill TW. Evidence that pp42, a major tyrosine kinase target protein, is a mitogen-activated serine/threonine kinase. Proc Natl Acad Sci U S A. 1989;86:6940-6943.
Boulton TG, Yancopoulos GD, Gregory JS, Slaughter C, Moomaw C, Hsu J, Cobb MH. An insulin stimulated protein kinase similar to yeast kinases involved in cell cycle control. Science. 1990;249:64-67.
Chen RH, Sarnecki C, Blenis J. Nuclear localization and regulation of erk- and rsk-encoded protein kinases. Mol Cell Biol. 1992;12:915-927.
Nel AE, Hanekom C, Rheeder A, Williams K, Pollack S, Katz R, Landreth GE. Stimulation of MAP-2 kinase activity in T lymphocytes by anti-CD3 or anti-Ti monoclonal antibody is partially dependent on protein kinase C. J Immunol. 1990;144:2683-2689.
Ferrell JE Jr, Martin GS. Platelet tyrosine-specific protein phosphorylation is regulated by thrombin. Mol Cell Biol. 1988;8:3603-3610.
Martinez R, Prevot-Mathey B, Bernards A, Baltimore D. Neuronal pp60c-src contains a six-amino acid insertion relative to its non-neuronal counterpart. Science. 1987;237:411-415.
Di Salvo J, Gifford D, Kokkinakis A. ATP- and phosphorylated-mediated stimulation of pp60c-src kinase activity in extracts from vascular smooth muscle. J Biol Chem. 1989;264:10773-10778.
Childs TJ, Watson MH, Sanghera JS, Campbell DL, Pelech SL, Mak AS. Phosphorylation of smooth muscle caldesmon by mitogen-activated protein (MAP) kinase and expression of MAP kinase in differentiated smooth muscle cells. J Biol Chem. 1992;267:22853-22859.
Collins EM, Walsh MP, Morgan KG. Contraction of single vascular smooth muscle cells by phenylephrine at constant [Ca2+]i. Am J Physiol. 1992;252:H754-H762.
Khalil RA, Lajoie C, Resnick MS, Morgan KG. Ca2+-independent isoforms of protein kinase C differentially translocate in smooth muscle. Am J Physiol. 1992;263:C714-C719.
Khalil RA, Morgan KG. PKC-mediated redistribution of mitogen-activated protein kinase during smooth muscle activation. Am J Physiol. 1993;265:C406-C411.
Agard DA, Hiraoka Y, Shaw P, Sedat JW. Fluorescence microscopy in three dimensions. In: Taylor DL, Wang Y, eds. Fluorescence Microscopy of Living Cells in Culture. New York, NY: Academic Press Inc; 1989:353-377.
Khalil RA, Morgan KG. Imaging of protein kinase C distribution and translocation in living vascular smooth muscle cells. Circ Res. 1991;69:1626-1631.
Khalil RA, Lajoie C, Morgan KG. In situ determination of [Ca2+]i threshold for translocation of the α-protein kinase C isoform. Am J Physiol. 1994;266:C1544-C1551.
Godt RE, Kirby AC, Gordon AM. Effects of hypertonic solutions on contraction of frog tonic muscle fibers. Am J Physiol. 1984;246:C148-C153.
Cecchi G, Bagni MA. Myofilament lattice spacing affects tension in striated muscle. News Physiol Sci. 1994;9:3-7.
Parker CA, Takahashi K, Tao T, Morgan KG. Agonist-induced redistribution of calponin in contractile vascular smooth muscle cells. Am J Physiol. 1994;267:C1262-C1270.
Yaish P, Gazit A, Gilon C, Levitzki A. Blocking of EGF-dependent cell proliferation by EGF receptor kinase inhibitors. Science. 1988;242:933-935.
Lyall RM, Zilberstein A, Gazit A, Gilon C, Levitzki A, Schlessinger J. Tyrphostins inhibit epidermal growth factor (EGF)-receptor tyrosine kinase activity in living cells and EGF-stimulated cell proliferation. J Biol Chem. 1989;264:14503-14509.
Shiraishi T, Owada K, Tatsuka M, Yamashita T, Watanabe K, Kakunaga T. Specific inhibitors of tyrosine-specific protein kinases: properties of 4-hydroxycinnamide derivatives in vitro. Cancer Res. 1989;49:2374-2378.
Koch CA, Anderson D, Moran MF, Ellis C, Pawson T. SH2 and SH3 domains: elements that control interactions of cytoplasmic signaling proteins. Science. 1991;252:668-674.
Lenormand P, Sardet C, Pages G, L’Allemain G, Brunet A, Pouyssegur J. Growth factors induce nuclear translocation of MAP kinases (p42mapk and p44mapk) but not of their activator MAP kinase kinase (p45mapkk) in fibroblasts. J Cell Biol. 1993;122:1079-1088.
Gonzalez FA, Seth A, Raden DL, Bowman DS, Fay FS, Davis RJ. Serum-induced translocation of mitogen-activated protein kinase to the cell surface ruffling membrane and the nucleus. J Cell Biol. 1993;122:1089-1101.
Sobue K, Sellers JR. Caldesmon, a novel regulatory protein in smooth muscle and nonmuscle actomyosin systems. J Biol Chem. 1991;266:12115-12118.
Wang C-LA, Wang L-WC, Xu S, Lu RC, Saavedra-Alanis V, Bryan J. Localization of the calmodulin- and the actin-binding sites of caldesmon. J Biol Chem. 1991;266:9166-9172.
Katsuyama H, Wang C-LA, Morgan KG. Regulation of vascular smooth muscle tone by caldesmon. J Biol Chem. 1992;267:14555-14558.
North AJ, Gimona M, Cross RA, Small JV. Calponin is localized in both the contractile apparatus and the cytoskeleton of smooth muscle cells. J Cell Sci. 1994;107:437-444.
Adam LP, Haeberle JR, Hathaway DR. Phosphorylation of caldesmon in arterial smooth muscle. J Biol Chem. 1989;264:7698-7703.