Increased Expression of Axl Tyrosine Kinase After Vascular Injury and Regulation by G Protein–Coupled Receptor Agonists in Rats
Abstract—Axl is a receptor tyrosine kinase originally identified as a transforming gene product in human myeloid leukemia cells. Cultured rat vascular smooth muscle cells also express Axl, where it has been proposed that Axl may play a role in cell proliferation. In the current study, we tested the hypotheses that Axl expression would parallel neointima formation in balloon-injured rat carotid, and that Axl expression would be regulated by growth factors present at sites of vascular injury. Ribonuclease protection assay showed dynamic increases in Axl mRNA in vessels, with peak expression 7 and 14 days after injury. Immunohistochemical analysis confirmed these results and demonstrated that Axl protein expression was localized primarily to cells of the neointima after injury. Northern blot analysis indicated increased mRNA expression for the secreted Axl ligand, Gas6, in injured carotids, with a time course paralleling that of Axl upregulation. Axl and Gas6 expression were temporally correlated with neointima formation, suggesting a role for Axl signaling in this process. Other studies, performed in cultured rat vascular smooth muscle cells, revealed positive regulation of Axl mRNA expression by thrombin or angiotensin II but not by basic fibroblast growth factor, platelet-derived growth factor-BB, or transforming growth factor-β1. Western blot analysis confirmed these results, showing that Axl protein expression was specifically increased by thrombin or angiotensin II. Our results implicate Axl as a potential mediator of vascular smooth muscle migration and proliferation caused by vascular injury and G protein–coupled receptor agonists.
Vascular neointima formation is a critical component of the reparative response to vessel injury. After injury, VSMCs proliferate and migrate in response to a dynamic and highly ordered program of expression of growth factors, receptors, and intracellular mediators. The rat carotid balloon-injury model has yielded major insights into the molecular mediators of the response to vascular injury. The characteristic response is manifested in sequential waves, encompassing replication of medial vascular smooth muscle cells (VSMCs) during the first 3 days, migration of VSMCs from media into the intimal space from days 3 to 14, and proliferation of VSMCs within the neointima peaking 7 to 14 days after injury.1 Studies of the rat carotid balloon-injury model suggest that intracellular signal transduction through tyrosine kinase cascades is a critical event in VSMC proliferation and migration. Expression of growth factors that signal through receptor tyrosine kinases, such as basic fibroblast growth factor (bFGF) and platelet-derived growth factor (PDGF), is increased in injured rat blood vessels, and these factors participate selectively in the 3 waves of vessel response to injury.1 Furthermore, inhibiting tyrosine kinases attenuates neointima formation.2
In addition to “classical” growth factors, the response to vessel injury is also regulated by vasoactive agonists that bind to G protein–coupled receptors, such as angiotensin II (Ang II), endothelin, and thrombin. Synthesis of these agonists and their receptors is dynamically regulated after vessel injury, and local regulation of the vascular renin-angiotensin system has been particularly well documented.3 4 5 6 7 These studies have indicated a major role for Ang II in neointimal proliferation at 7 to 14 days post–balloon catheter injury. Like classical growth factors, Ang II and thrombin increase cellular protein tyrosine phosphorylation in VSMCs, but the identities of the kinases responsible remain largely unknown.8 Our laboratory has had a long-standing interest in the identification and characterization of tyrosine kinases involved in VSMC proliferation.9
Axl (also called UFO or Ark) is a 140-kD protein that was originally identified in human patients with chronic myelogenous leukemia.10 11 The transforming ability of axl was subsequently found to be due to overexpression of the gene by tumor cells.10 A role for Axl in cardiovascular pathophysiology has only recently been suggested. Nakano et al12 reported that stimulation of Axl by its ligand, Gas6,13 14 15 increased cultured VSMC DNA synthesis when coadministered with Ang II, thrombin, or lysophosphatidic acid. This stimulation of DNA synthesis was much greater than that caused by Gas6 or these agonists alone. Based on these results, we hypothesized that Axl would be dynamically regulated during vascular growth and neointima formation. This hypothesis was investigated in the present study by determining the time course of Axl expression after balloon catheter injury in the rat carotid. To define the agonists that regulate Axl during vascular injury, we also tested the effects of Ang II, thrombin, transforming growth factor-β (TGF-β), PDGF, and bFGF on Axl expression. The data in this study provide strong evidence that Axl expression is regulated by G protein–coupled receptor agonists and may be involved in the vascular response to injury.
Materials and Methods
Rat Carotid Injury
Male Sprague-Dawley rats (400 to 450 g, Charles River Laboratories, Wilmington, Mass) anesthetized with ketamine (80 mg/kg, IM; Fort Dodge) and xylazine (5 mg/kg, IM; Phoenix Pharmaceuticals) were used. All procedures were carried out in a specific pathogen-free animal care facility at the University of Washington and were approved by the University of Washington Animal Use and Care Committee. The left carotid artery was isolated via a ventral neck incision. The external carotid was ligated at the level of the hyoid bone and a ligature was placed directly above the bifurcation. A cut was made between the ligatures and the artery cannulated with a 2F Fogarty embolectomy catheter (Baxter). The catheter was advanced 5 cm through the common carotid and the balloon was inflated to yield light resistance. The catheter was then pulled, with twisting, through the artery, maintaining resistance. This procedure was repeated twice more, and the external carotid was ligated. Rats received nalbuphine hydrochloride (2.5 mg/kg, SC, Abbott Laboratories) for analgesia, and ticarcillin (100 mg/kg, IM, SmithKline Beecham) was given to prevent infection. Sham-operated rats were treated identically, except that the catheter was not passed through the carotid. At the end of the experiment, rats were anesthetized with pentobarbital sodium (50 mg/kg, IP, Abbott Laboratories), and the carotids were removed. Animals whose vessels were used for immunohistochemical studies were perfusion-fixed at 130 mm Hg with 10% formalin. Vessels used for total RNA extraction were removed from the anesthetized rats and snap-frozen in liquid nitrogen.
Cloning of Rat Axl Partial cDNA
Degenerate polymerase chain reaction (PCR) was used to clone Axl from rat carotid 4 days after balloon injury. Total RNA (5 μg, harvested as below) was used as template for cDNA synthesis, using reverse transcriptase. PCR primers were targeted to highly conserved tyrosine kinase domains 6 and 9 (upper primer: GGAATTCCYKNRTNCAYMGIGAYHTIGCIGYIMGIAAY; lower primer: CGGGATCCCGCCAIADIVDIAYISCRWAIVHCCAIACRT). PCR conditions were as follows: 40 μmol/L each dATP, dCTP, dTTP, dGTP, 1 μL 35S-ATP (1200 mCi/mL, NEN), and Taq polymerase (10 U; Perkin-Elmer). Cycle parameters were 94°C for 2.5 minutes; 5 cycles of 94°C, 50°C, 72°C each for 30 seconds; and 35 cycles of 94°C, 56°C, 72°C each for 30 seconds. The reaction product obtained was separated on a 6% sequencing gel (National Diagnostics), visualized on an autoradiogram, cut from the gel, and reamplified using the same PCR primers. The PCR product was restricted with BamHI and EcoRI and ligated into pGEM 3Zf(+) (Promega). After transformation of E coli, the product was bidirectionally sequenced and found to be 91% identical and 98% homologous at the amino acid level to both human and mouse Axl. The sequence of the rat Axl partial cDNA is: AATTCCCTGGTGCACAGGGATTTG- GCGGCGCGGAACTGCATGCTGAATGAGAACATGTCCGTG- TGCGTGGCAGACTTCGGGCTCTCCAAGAAGATCTACAATG- GGGATTACTACCGCCAAGGGCGCATTGCCAAGATGCCAG- TCAAGTGGATTGCTATCGAGAGTCTGGCAGATCGAGTCTA- ACACCAGCAAGAGTGACGTCTGGGCCTACGCCATCACCAT- CTTGGGCGGGATCC. This sequence has been submitted to Gen- Bank (accession number AF046886).
Ribonuclease Protection Assay
Total RNA was isolated from frozen tissue or cultured cells using a commercially-available kit (Totally RNA, Ambion).32P radiolabeled ([α-32P]UTP [800 Ci/mmol], New England Nuclear) antisense riboprobes were synthesized using partial cDNAs to rat Axl (235 bp) and rat GAPDH (316 bp; Ambion). Riboprobe synthesis conditions were as follows: 50 μCi [α-32P]UTP, 125 μmol/L each rATP, rCTP, rGTP (GIBCO-BRL), 5 μmol/L UTP (GIBCO-BRL), 10 mmol/L DTT (GIBCO-BRL), and 40 U RNase inhibitor (Ambion) in a volume of 20 μL. Each probe (100 000 cpm [1 μL]) was hybridized to 10 μg total RNA and ribonuclease protection assay (RPA) was performed using a commercially available kit (HybSpeed RPA, Ambion). Samples were then run on a 6% sequencing gel (National Diagnostics) and subjected to autoradiography. Densitometry was performed with a LaCie light scanner and NIH Image, version 1.59, was used for analysis. The ratio of Axl density relative to GAPDH (internal control) density was derived for each experimental time point. The ratio of the sham or unstimulated condition was arbitrarily set to 1.0 for each experiment to allow statistical comparison among the different experiments.
Paraffin embedded rat carotids (5-μm-thick sections) were deparaffinized and incubated in 80% methanol containing 0.6% hydrogen peroxide for 30 minutes to quench the endogenous peroxidases and blocked with 5% normal horse serum for 30 minutes. All solutions were bought from Vector unless otherwise specified and were prepared in PBS with Ca2+ and Mg2+ containing 1% bovine serum albumin (Sigma) and applied at room temperature. There were at least three 5- to 10-minute washes between each solution application. All washes were done in PBS. Mouse monoclonal antibody for Axl (1:250 dilution; Transduction Laboratories) was used. This monoclonal antibody was compared with the polyclonal anti-Axl used for Western blotting (see below) for specificity. The monoclonal antibody revealed the same pattern of bands as the polyclonal antibody on Western blots of lysates from A431 cells, a cell line which abundantly expresses Axl (data not shown). However, when tested on fixed tissue sections for immunohistochemistry, the monoclonal antibody gave a lower background. The antibody was applied in pools, and sections were incubated for 1 hour in a humidifier box. Biotinylated secondary antibody made in horse against mouse (1:500) was applied for 1 hour, followed by a 30-minute incubation in ABC and 5-minute development in 0.5% DAB in 50 mmol/L Tris at pH 7.6. For a negative control, primary antibody was substituted with normal mouse IgG at a corresponding dilution. The cross sections were counterstained with hematoxylin.
Vessels were harvested from anesthetized rats after perfusion with lactated Ringer’s solution (Baxter) and immediately placed in ice-cold PBS containing 10 μg/mL leupeptin and 10 μg/mL soybean trypsin inhibitor. The vessels were then pinned and cut longitudinally in ice-cold PBS containing protease inhibitors under a dissecting microscope. The endothelium was removed by manual scraping, and the medial layer was rapidly separated from the adventitia by peeling. The medial tissue was minced and proteins were extracted by boiling in 3× Laemmli sample buffer (150 mmol/L Tris, pH 6.8, 5% SDS, 2.5% β-mercaptoethanol, 0.02% bromophenol blue). The neointima was peeled away from the media and the 2 components were analyzed separately in vessels from 14-day injured rats. Lysates were then subjected to SDS-PAGE and immunoblotted as below.
Cultured cells were rinsed twice with 5 mL ice-cold PBS (140 mmol/L NaCl, 2.7 mmol/L KCl, 8 mmol/L Na2HPO4, and 1.5 mmol/L KH2PO4), and 1 mL hypotonic buffer (5 mmol/L HEPES, pH 7.4, 2 mmol/L MgCl2, 2.5 mmol/L DTT, 10 μg/mL leupeptin, and 0.1 mmol/L PMSF) was added to the plates. After 30 minutes’ incubation on ice, cells were scraped and lysed with 20 strokes in a Dounce homogenizer. Lysates were centrifuged at 900g for 5 minutes to pellet nuclei. The supernatants were then centrifuged for 30 minutes at 100 000g. The pellets, enriched in cell membranes, were resuspended in lysis buffer (1% Triton X-100, 50 mmol/L β-glycerophosphate, 200 mmol/L sodium orthovanadate, and 10 μg/mL leupeptin, in PBS) and placed on ice for 30 minutes. Cellular protein was then quantified (DC protein assay, Bio-Rad). Protein (25 to 50 μg) was subjected to SDS-PAGE and transferred to nitrocellulose (Amersham). Membranes were blotted with rabbit anti-Axl polyclonal antibody13 (1:1000 dilution; kindly provided by Dr Brian Varnum, Amgen, Thousand Oaks, CA), followed by incubation with donkey anti-rabbit IgG conjugated to horseradish peroxidase (1:1000 dilution; Amersham). Results were visualized with chemiluminescence (ECL, Amersham) and autoradiography. The anti-Axl polyclonal antibody identifies 2 to 3 bands on Western blots. The dominant band is ≈140 kD, and corresponds to mature, fully glycosylated Axl.16 Less prominent bands appear at ≈120 kD and ≈97 kD. All bands can be competed away by incubating the antibody with soluble Axl extracellular domain before Western blotting (data not shown). Based on these results, we believe that the lower-molecular-weight band represents immature, nonglycosylated Axl. To further characterize the Axl-immunoreactive bands, we measured agonist-induced Axl tyrosine phosphorylation. Cultured VSMCs were treated with Gas6 for 15 minutes and lysates were immunoprecipitated with PY99 anti-phosphotyrosine antibody (Santa Cruz). The immunoprecipitated proteins were size-fractionated on an SDS-PAGE gel, transferred to nitrocellulose, and immunoblotted with anti-Axl. Only the band at 140 kD appeared on the blot (data not shown). Thus, the lower band does not correspond to an Axl species that is tyrosine phosphorylated. Cell fractionation experiments revealed that the lower-molecular-weight species is present primarily in the cytosolic and cytoskeletal fractions (data not shown). Based on these 2 experiments, it is unlikely that the smaller Axl species is important in transmembrane signaling.
Common carotid arteries and thoracic aortae were harvested from normal rats, as well as from rats 6 hours and 3, 7, 14, and 28 days after balloon injury (3 to 4 animals per time point). Vessels were stripped of periadventitial fat and connective tissue in phosphate-buffered saline at 4°C and were then snap-frozen in liquid nitrogen. Frozen arterial tissue was ground to a fine powder under liquid nitrogen, and total cellular RNA was prepared by acid guanidinium thiocyanate extraction.17 Agarose gel electrophoresis of RNA (15 μg total RNA per lane) and transfer to nylon membranes (Zeta Probe, BioRad) were carried out as previously described.18 After transfer, RNA blots were exposed to shortwave UV light both to cross-link RNA to the membrane and to visualize the major ribosomal RNA bands. The blot was hybridized using cDNA probes labeled with [32P]dCTP by random primer extension (Amersham), washed at 65°C in 2 changes of 0.045 mol/L NaCl/0.0045 mol/L sodium citrate, pH 7.0/0.1% SDS for 20 minutes each, and then exposed to Kodak X-AR5 film at −70°C. A 500-bp cDNA clone encoding rat 28S rRNA and a 2400-bp cDNA clone encoding rat Gas6 (sequenced from a cDNA library from balloon-injured rat carotid) were used as templates for synthesis of [32P]dCTP-labeled probes.
Cultured rat VSMCs, prepared as described19 and obtained from frozen stocks, were used at passages 7 to 15. Cells were grown to near confluence in 100-mm dishes in DMEM (GIBCO-BRL) supplemented with 10% calf serum and penicillin/streptomycin. Forty-eight hours before agonist treatment, cells were growth-arrested by replacing the medium with DMEM containing 0.4% calf serum.
TGF-β1 was obtained from Boehringer Mannheim. Losartan was kindly provided by DuPont-Merck (Wilmington, Del). All other chemicals and reagents were purchased from Sigma.
Data are expressed as mean±SEM. Between-group comparisons were performed using 1-way ANOVA. Post hoc comparisons were done using Fisher’s least significant difference test with Systat for MacIntosh, version 5.1. Differences were considered significant if P<0.05.
RPA was used to examine the time course of Axl mRNA expression in rat carotid arteries after balloon injury. During the first 4 days, Axl mRNA expression (normalized to GAPDH mRNA expression) was not significantly different from the baseline mean of the sham-treated group (Figure 1⇓). Normalized Axl mRNA expression began to increase at 7 days (1.4±0.1-fold) and was significantly increased at 14 days (1.8±0.7-fold, P=0.017, n=3).
To determine which cells expressed Axl protein, immunohistochemical analysis was performed on cross sections of 4-, 7-, and 14-day injured carotids and carotids from sham-operated rats (Figure 2⇓). Abundant Axl protein expression was observed on staining with the anti-Axl antibody, localized mainly to the neointima (Figure 2A⇓ and 2F⇓) 14 days after injury. The neointima at 14 days showed no specific staining with normal rabbit serum (Figure 2B⇓). Sham rats (Figure 2C⇓) also showed only minimal staining in the media and adventitia. Examination of the immunohistochemical sections at high power (Figure 2C⇓ to 2F⇓) revealed a time-dependent increase in Axl protein, beginning between 4 and 7 days in the first patches of neointima cells (Figure 2D⇓ and 2E⇓). In agreement with the RPA data, increased Axl expression was localized mainly to the concentric neointima at 7 days (Figure 2E⇓) and most abundantly at 14 days (Figure 2F⇓), although the 7-day sections show some staining of medial cells. The results of the immunohistochemistry were confirmed by Western blotting of media and neointima from 14-day injured carotids and media of uninjured carotids (Figure 3⇓). While a very faint signal was seen in uninjured carotids, tissue taken from injured vessels was strongly positive for Axl. Furthermore, the distribution of Axl within the vessel wall 14 days after injury was identical to that seen by immunohistochemistry, with the majority of Axl expressed in the neointima. Such discrete localization of Axl expression, along with the normalization of expression relative to GAPDH, may explain the seemingly modest increase in Axl mRNA expression seen in whole injured vessels (Figure 1⇑). Figure 2⇓ would predict that many medial and neointima cells expressed GAPDH (a ubiquitous “housekeeping” gene) but not appreciable amounts of Axl. Thus, Axl transcripts present in total vessel RNA may have been “diluted” by RNA from nonexpressing cells.
Several groups have reported that Gas6, a protein secreted by rat VSMCs,12 is an agonist for Axl, resulting in autophosphorylation of the receptor.13 14 15 We used Northern blotting to examine the expression of Gas6 mRNA in balloon-injured rat carotid, in an effort to confirm presence of the Axl ligand in vivo. Gas6 expression (normalized to 28S RNA) was relatively low in normal vessels and 6 hours postinjury (Figure 4⇓). Three days after injury, expression was increased ≈2-fold. Gas6 expression was maximal 7 days after injury (≈3-fold increase), but remained above baseline at all later time points tested (2 to 4 weeks). Thus, the Axl ligand is upregulated in rat carotid after balloon injury, suggesting increased activity of the Axl-Gas6 system.
It is well established that many growth factors and vasoactive agents present at sites of vascular injury can act as mitogens for VSMCs.1 To identify which agents may positively regulate Axl expression, RPA was performed on total RNA from cultured VSMCs treated with thrombin, Ang II, bFGF, PDGF-BB, or TGF-β1. The concentrations of agonists used were chosen after review of the literature as those causing maximal receptor activation in tissue-culture systems. As shown in Figure 5⇓, thrombin (10 U/mL) increased normalized Axl mRNA expression significantly at 2 hours (3.0±1.2-fold), 4 hours (3.2±0.4-fold, P<0.05, n=4), and 8 hours (2.7±0.4-fold). Ang II (1 μmol/L) caused a large, rapid, increase in Axl mRNA expression in VSMCs (Figure 6⇓). At 1 hour, Axl expression increased 4.7±1.3-fold compared with untreated cells. Axl mRNA expression was increased 6.0±1.7-fold at 4 hours (P<0.05, n=6) and was still increased 3.1±1.1-fold after 24 hours of Ang II treatment. The effect of Ang II on Axl mRNA expression was mediated by the AT1 receptor, as concomitant losartan treatment (10 μmol/L) completely blocked Ang II–induced Axl mRNA induction at all time points.
Other growth factors involved in the vascular response to injury include bFGF, PDGF-BB, and TGF-β1.1 Stimulation of cultured VSMCs with 20 ng/mL bFGF decreased Axl mRNA expression, whereas 10 ng/mL PDGF-BB and 2 ng/mL TGF-β1 had minimal, nonsignificant effects on Axl mRNA expression compared with either thrombin or Ang II (Figure 7⇓). These results, when combined with the results shown in Figures 5⇑ and 6⇑, suggest that Axl regulation in VSMCs in vitro is a specific property of G protein–coupled receptor agonists.
Western blotting was performed to determine the time course of Axl protein expression after treatment of VSMCs with Ang II, thrombin, bFGF, or PDGF-BB. Ang II and thrombin, but not bFGF or PDGF-BB, increased expression primarily of a 140-kD protein that was immunoreactive with Axl antibody (maximal 11-fold and 4-fold over untreated cells, respectively; Figure 8⇓). The most prominent band observed was near the molecular weight previously reported for Axl, indicating that the antibody detected its mature (glycosylated) form.10 In contrast to the positive control A431 cells, the rat VSMC protein was slightly smaller, suggesting cell-specific differences in the extent of glycosylation. The time course of Axl protein induction by Ang II and thrombin is consistent with the effect of these agonists on Axl mRNA induction (Figures 5⇑ and 6⇑). In agreement with the RPA data, bFGF and PDGF did not significantly increase in Axl protein expression after treatment with these growth factors (Figure 8⇓).
In the current study, we have demonstrated increased expression of Axl, a recently described receptor tyrosine kinase, and its ligand, Gas6, in rat carotid arteries after balloon injury. This is the first report of in vivo Axl regulation in the cardiovascular system. Furthermore, we have shown positive regulation of Axl mRNA and protein levels in cultured rat VSMCs by growth factors, specifically the G protein–coupled receptor agonists thrombin and Ang II. Our results suggest important interactions between Ang II (or thrombin) and Axl. Thus, future characterization of Axl signaling should yield important discoveries regarding the mechanisms underlying the response of VSMCs to vascular injury.
The time course of Axl expression in injured rat carotid, which was maximal at 7 to 14 days, is consistent with the hypothesis that Ang II is an important regulator of Axl expression in vivo. Support for this assertion comes from several studies that have examined the renin-angiotensin system in the rat balloon-injury model. Viswanathan et al6 7 found that AT1 receptor expression was increased approximately 4-fold in neointima of injured aortae or carotids 8 to 15 days postinjury. In addition, these investigators reported much greater expression of the receptor in neointima compared with medial smooth muscle cells. AT1 receptor expression was enhanced in smooth muscle cells immunoreactive for proliferating cell nuclear antigen,7 suggesting a role for Ang II in neointima cell proliferation. Similar results were obtained by deBlois et al,3 who reported that Ang II–induced DNA synthesis correlated with increased expression of AT1 receptor in the neointima in the early weeks after injury. Angiotensin-converting enzyme expression and activity have also been shown to be increased after balloon injury, with a time course (7 to 14 days) and distribution within the vessel wall (greatest in neointima) similar to the changes which occur in AT1 receptor levels.4 Angiotensinogen is also increased in neointima.5 Prescott et al20 reported that Ang II contributes to neointima formation by causing both migration and proliferation of medial VSMCs in the injured rat carotid and that a large effect of Ang II on neointima lesion size was apparent 12 days postinjury. Taken together, these data are supportive of an important role for Ang II in growth after vascular injury. In the present study, RPA showed that Axl mRNA expression was low until 7 to 14 days postinjury. Our immunohistochemical and Western blotting studies revealed the same time course. Therefore, Axl expression after vascular injury in vivo is temporally and spatially correlated with neointima formation and is similar to the distribution of angiotensin-converting enzyme.4 Overall, these results are consistent with the hypothesis that Ang II generated within the subluminal neointima cells induces Axl expression in vivo.
Our finding of increased Axl expression in VSMCs within the neointima of balloon-injured rat carotid arteries is especially intriguing in light of the data published by Nakano et al.12 Using cultured rat VSMCs, these investigators reported that Gas6, an autocrine ligand for Axl, increased DNA synthesis only when coadministered with thrombin, Ang II, or lysophosphatidic acid. Thus, it appears that Axl-mediated VSMC growth can occur only in the presence of these agents. Since the thrombin and Ang II signaling pathways have been shown to be upregulated in proliferating VSMCs,1 it is possible that Axl may be responsible in part for the VSMC proliferation characteristic of neointima formation. In addition, we show in the current study that Gas6 expression is increased in rat carotid after balloon injury (Figure 4⇑), further supporting the hypothesis that Axl plays an important role in neointima formation. The humoral regulators of Gas6 expression in VSMCs have not been determined to date, but identification of such factors is an important goal for future studies.
The in vitro studies we performed in cultured VSMCs yielded important insights into the regulation of Axl expression. Of the agonists tested, Ang II, and to a lesser extent, thrombin, increased Axl mRNA and protein expression. In contrast, the effects of PDGF-BB and TGF-β1 on Axl expression in vitro were markedly less than those of the G protein–coupled receptor agonists tested. Surprisingly, bFGF decreased Axl expression by unknown mechanisms. Our in vitro results with Ang II lend support to the assertion that the peptide can regulate Axl expression in vivo. The mitogenic activity of Ang II is known to be dependent on the AT1 receptor in VSMCs from adult animals.3 21 Our finding that the AT1 receptor antagonist losartan completely blocked the stimulatory effect of Ang II on Axl mRNA and protein expression (Figure 6⇑ and data not shown, respectively) is consistent with the hypothesis that upregulation of Axl expression is one of the mechanisms through which Ang II enhances VSMC proliferation.
The finding of selective regulation of Axl by G protein–coupled receptor agonists is interesting for several reasons. First, the data suggest that the downstream nuclear targets of the AT1 and thrombin receptors differ from those of classical growth factor receptors. The transcription factors activated by the AT1 and thrombin receptors are likely specific in regard to the axl gene, despite the fact that these receptors, along with the PDGFβ and FGF receptors, increase AP-1 and STAT activity.8 Second, only situations marked by increased Ang II (or thrombin) generation would be expected to result in Axl-mediated cell proliferation. This could have important implications for the vascular and cardiac hypertrophy associated with Ang II–dependent hypertension.22 23 24 25 Third, the present results provide another pathway by which activation of G protein–coupled receptors leads to increased cellular protein tyrosine phosphorylation.
In summary, we have shown that expression of the recently described receptor tyrosine kinase Axl is increased in neointima VSMCs of rat carotid arteries after balloon injury and that Axl expression is selectively regulated by G protein–coupled receptor agonists in vitro. Our results suggest that Axl may play a role in VSMC proliferation after vascular injury and exposure to Ang II or thrombin. Axl may be one of the mediators of the increased cellular protein tyrosine phosphorylation that occurs upon Ang II or thrombin stimulation of VSMCs. Future experiments will focus on molecular dissection of the interaction between Axl and G protein–coupled receptors.
This work was supported by NRSA 1F32HL-09780 (Matthew G. Melaragno), NRSA 5F32HL-09201 (Daniel A. Wuthrich), NIH R01HL-49192 (Bradford C. Berk, Veronica Poppa, and Denzil Gill), and NIH 1K08HL-03282 (Marshall A. Corson). Bradford C. Berk and Volkhard Lindner are Established Investigators of the American Heart Association. Rabbit anti-Axl polyclonal antibody was kindly provided by Dr Brian Varnum, Amgen, Thousand Oaks, Calif, and losartan by DuPont-Merck, Wilmington, Del. The experimental assistance of Sara L. MacKellar is gratefully acknowledged.
This manuscript was sent to Valentin Fuster, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
- Received November 24, 1997.
- Accepted July 7, 1998.
- © 1998 American Heart Association, Inc.
Schwartz SM, deBlois D, O’Brien ER. The intima: soil for atherosclerosis and restenosis. Circ Res. 1995;77:445–465.
deBlois D, Viswanathan M, Su JE, Clowes AW, Saavedra JM, Schwartz SM. Smooth muscle DNA replication in response to angiotensin II is regulated differently in the neointima and media at different times after balloon injury in the rat carotid artery. Role of AT1 receptor expression. Arterioscler Thromb Vasc Biol. 1996;16:1130–1137.
Fishel RS, Thourani V, Eisenberg SJ, Shai SY, Corson MA, Nabel EG, Bernstein KE, Berk BC. Fibroblast growth factor stimulates angiotensin converting enzyme expression in vascular smooth muscle cells. J Clin Invest. 1995;95:377–387.
Rakugi H, Jacob HJ, Krieger JE, Ingelfinger JR, Pratt RE. Vascular injury induces angiotensinogen gene expression in the media and neointima. Circulation. 1993;87:283–290.
Viswanathan M, Stromberg C, Seltzer A, Saavedra JM. Balloon angioplasty enhances the expression of angiotensin II AT1 receptors in neointima of rat aorta. J Clin Invest. 1992;90:1707–1712.
Corson MA, Berk BC. Tyrosine kinases in the regulation of vascular smooth muscle function. In: Marks AR, Taubman MB, eds. Molecular Biology of Cardiovascular Disease. New York, NY: Marcel Dekker, Inc; 1997:291–325.
Berk BC, Corson MA. Angiotensin II signal transduction in vascular smooth muscle: role of tyrosine kinases. Circ Res. 1997;80:607–616.
O’Bryan JP, Frye RA, Cogswell PC, Neubauer A, Kitch B, Prokop C, Espinosa RD, Le Beau MM, Earp HS, Liu ET. Axl, a transforming gene isolated from primary human myeloid leukemia cells, encodes a novel receptor tyrosine kinase. Mol Cell Biol. 1991;11:5016–5031.
Nakano T, Higashino K, Kikuchi N, Kishino J, Nomura K, Fujita H, Ohara O, Arita H. Vascular smooth muscle cell-derived, Gla-containing growth-potentiating factor for Ca2+-mobilizing growth factors. J Biol Chem. 1995;270:5702–5705.
Nagata K, Ohashi K, Nakano T, Arita H, Zong C, Hanafusa H, Mizuno K. Identification of the product of growth arrest-specific gene 6 as a common ligand for Axl, Sky, and Mer receptor tyrosine kinases. J Biol Chem. 1996;271:30022–30027.
O’Bryan JP, Fridell YW, Koski R, Varnum B, Liu ET. The transforming receptor tyrosine kinase, Axl, is post-translationally regulated by proteolytic cleavage. J Biol Chem. 1995;270:551–557.
Majesky MW, Reidy MA, Bowen-Pope DF, Hart CE, Wilcox JN, Schwartz SM. PDGF ligand and receptor gene expression during repair of arterial injury. J Cell Biol. 1990;111:2149–2158.
Duff JL, Marrero MB, Paxton WG, Charles CH, Lau LF, Bernstein KE, Berk BC. Angiotensin II induces 3CH134, a protein-tyrosine phosphatase, in vascular smooth muscle cells. J Biol Chem. 1993;268:26037–26040.
Weber H, Taylor DS, Molloy CJ. Angiotensin II induces delayed mitogenesis and cellular proliferation in rat aortic smooth muscle cells: correlation with the expression of specific endogenous growth factors and reversal by suramin. J Clin Invest. 1994;93:788–798.
Lever AF. The fast and slowly developing pressor effect of angiotensin II. In: Robertson JIS, Nicholls MG, eds. The Renin-Angiotensin System. London, UK: Gower Medical Publishing; 1993:28.1–28.9.
Griffin SA, Brown WC, MacPherson F, McGrath JC, Wilson VG, Korsgaard N, Mulvany MJ, Lever AF. Angiotensin II causes vascular hypertrophy in part by a nonpressor mechanism. Hypertension. 1991;17:626–635.
Robertson J. Renin and the pathophysiology of renovascular hypertension. In: Robertson JIS, Nicholls MG, eds. The Renin-Angiotensin System. London, UK: Gower Medical Publishing; 1993:55.1–55.34.
Capers Q, Laursen JB, Fukui T, Rajagopalan S, Mori I, Lou P, Freeman BA, Berrington WR, Griendling KK, Harrison DG. Vascular thrombin receptor regulation in hypertensive rats. Circ Res. 1997;80:838–844.