Membrane-Bound Protein Kinase A Inhibits Smooth Muscle Cell Proliferation In Vitro and In Vivo by Amplifying cAMP–Protein Kinase A Signals
Abstract—cAMP-dependent protein kinase is anchored to discrete cellular compartments by a family of proteins, the A-kinase anchor proteins (AKAPs). We have investigated in vivo and in vitro the biological effects of the expression of a prototypic member of the family, AKAP75, on smooth muscle cells. In vitro expression of AKAP75 in smooth muscle cells stimulated cAMP-induced transcription, increased the levels of the cyclin-dependent kinase-2 inhibitor p27kip1, and reduced cell proliferation. In vivo expression of exogenous AKAP75 in common carotid arteries, subjected to balloon injury, significantly increased the levels of p27kip1 and inhibited neointimal hyperplasia. Both the effects in smooth muscle cells in vitro and in carotid arteries in vivo were specifically dependent on the amplification of cAMP-dependent protein kinase (PKA) signals by membrane-bound PKA, as indicated by selective loss of the AKAP75 biological effects in mutants defective in the PKA anchor domain or by suppression of AKAP effects by the PKA-specific protein kinase inhibitor. These data indicate that AKAP proteins selectively amplify cAMP-PKA signaling in vitro and in vivo and suggest a possible target for the inhibition of the neointimal hyperplasia after vascular injury.
Cyclic AMP controls growth and differentiation in a variety of organisms and cell types.1 2 In eukaryotes, cAMP binds the regulatory subunit of cAMP-dependent protein kinases (PKAs). This releases the catalytic subunit (C-PKA), which phosphorylates a wide variety of substrate proteins. A fraction of the C-PKA migrates to the nucleus and phosphorylates nuclear proteins and transacting factors.3
PKA is targeted to certain subcellular locations by specific anchor proteins (A-kinase anchor proteins, AKAPs). Localized PKA holoenzymes might in vivo perform important aspects of cAMP-activated signal transduction. We have previously provided evidence that links membrane targeting of PKAII to cAMP-dependent gene transcription in differentiated and nondifferentiated cells.4 5 In thyroid, neuronal, and kidney cells, displacement of immobilized PKAII from perinuclear sites to the cytoplasm impaired cAMP-induced transcription.4 Conversely, overexpression of AKAP75, a prototype AKAP that targets PKA to the membranes, enhanced the propagation of cAMP signals to the nucleus.5
However, the biological effects of AKAPs on smooth muscle cell (SMC) growth in vitro and in vivo are not known. Therefore, we performed 2 experimental models to assess the effects of AKAP expression on SMC proliferation both in vitro and in vivo. In the first model, we have transfected stabilized SMCs derived from aorta with expression vectors encoding the prototypic member of AKAP family, AKAP75 and its derivatives. The same vectors were also used in vivo in the rat model of angioplasty. After balloon injury, the cells in the media of the arterial wall are potently stimulated and proliferate vigorously, thus generating the so-called neointima.6 This system is ideally suited for assessing nontransformed SMC proliferation in vivo7 8 9 10 and testing innovative strategies to prevent vascular proliferative disorders such as atherosclerosis and restenosis.
Here we report that overexpression in vitro and in vivo of AKAP75 potently inhibits SMC proliferation and prevents neointima formation after balloon injury. These effects are specifically dependent on the amplification of cAMP signals by membrane-bound PKA and suggest a potential target site for therapeutic control of balloon angioplasty–induced neointimal formation.
Materials and Methods
DNA Plasmids, Cell Culture, and Transfections
The plasmids used in this study are (1) cAMP-responsive element (CRE)-chloramphenicol acetyltransferase (CAT), carrying the Escherichia coli CAT gene under the control of the somatostatin promoter11 ; (2) CMV-AKAP75, containing the bovine brain RIIβ anchor protein under the control of cytomegalovirus (CMV) promoter12 ; (3) CMV-AKAP45, a truncated mutant of AKAP75 and a derivative of AKAP45 mutated in the RII binding site (AKAP45mut)12 ; (4) RSV-LacZ, the E. coli LacZ gene driven by the long terminal repeats of Rous sarcoma virus (RSV)13 ; (5) RSV-PKI,14 containing the RSV promoter driving the expression of the rabbit PKA inhibitor gene; (6) PKImut, the rabbit PKA inhibitor single-point mutant gene14 ; (7) RSV-Neo, the neomycin resistance gene; and (8) RSV-CAT, which carries the CAT gene under the control of RSV promoter.
Rat aortic embryonal SMCs (A10 cells) were cultured in DMEM (10% FCS), and DNA transfections were carried out by the calcium phosphate procedure.11 For stable transformed lines (CMV-AKAP75, RSV-Neo, and AKAP45mut), after 24 hours of incubation in nonselective medium, the cells were plated in the selective medium with a neomycin analogue (G418, 500 μg/mL).
Dishes were transfected with 5 μg each of CRE-CAT, RSV-Lac Z, AKAP75, AKAP45, AKAP45mut, and protein kinase inhibitor (PKI), as well as up to 21 μg of salmon sperm DNA.4 CAT activity was performed 48 hours after transfection before and after 4 hours of 8-bromo-cAMP (8-Br-cAMP) incubation.11
To assess the effects of AKAP75 transfection on SMC proliferation in vitro, the cultures were synchronized by starvation for 48 hours in serum-free DMEM and then serum-restimulated in the presence of [3H]thymidine. DNA synthesis cells were measured by [3H]thymidine incorporation as described previously.12 Experiments were performed in quadruplicate, and the data shown are a mean of 4 independent experiments that all gave similar results.
Western Immunoblot Analysis
Dishes were transfected with 5 μg each of RSV-Lac Z, AKAP75, and PKI. Western immunoblot analysis was performed 48 hours after transfection before and after 4 hours of 8-Br-cAMP incubation.12 In brief, after cell homogenization, proteins were transferred to a nitrocellulose membrane (Hybond-ECL, Amersham Life Sciences) by semidry electroblotting for 1 hour. The membranes were blocked for 1 hour at room temperature with Blotto-Tween (5% nonfat dry milk, 0.1% Tween-20) and incubated with a murine polyclonal p27kip1 IgG1 (0.25 mg/mL, dilution 1:1000, Santa Cruz Biotechnology, Inc). Bound antibody was detected with horseradish peroxidase–labeled rabbit anti-mouse IgG conjugate (Prosan, dilution 1:2000 in Blotto-Tween) and visualized by enhanced chemiluminescence (Amersham).
Ligand Blot Analysis or Overlay
To identify AKAP75-RIIβ complex, after electrophoresis, the proteins were transferred to nitrocellulose filters, probed with labeled RIIβ (AKAP75 antibody was kindly provided from Dr CS Rubin (Albert Einstein College of Medicine, Bronx, NY), and identified by autoradiography as previously described.11
All animals received humane care in accordance with the animal use principles of the American Society of Physiology. In 64 Wistar rats (350 to 400 g), balloon injury of the common carotid artery was performed as previously described.7 8 9 Site-specific gene transfer was achieved by transfecting arteries with the use of a DNA/gel (30% wt/vol) solution (pluronic gel F127, BASF Corp) containing different plasmids (200 μg each plasmid), applied on the artery immediately after the injury.8 9
We assessed the effects of balloon injury (n=11) alone and of pluronic gel application (without plasmids) (n=6) after the injury, on neointimal formation 14 days after injury.11 Then we determined the effect of the following plasmid constructs: (1) RSV–β-Gal (n=11); (2) CMV-AKAP75 (n=12); (3) CMV-AKAP45 (n=7); and (4) finally, the effect of simultaneous transfection of AKAP75 with PKI15 (n=5) or PKImut (n=4).
Localization of AKAP75 and p27kip1 in the Vessel Wall
Two days after balloon injury, additional animals from the control and AKAP75 groups were euthanized by administration of an overdose of pentobarbital, and the arteries were perfusion-fixed in 4% formaldehyde.16 Arteries were divided into 2-mm-thick segments, overlaid with OCT compound, and frozen in liquid nitrogen. Cryostat sections (7 μm) were mounted on poly-l-lysine–coated slides. To detect AKAP75 and p27kip1 immunoreactivity, sections were washed twice with Tris-buffered saline (TBS), blocked with rabbit serum, diluted 1:5 in TBS for 45 minutes, incubated overnight with polyclonal murine anti-AKAP75 or anti-p27kip1 IgG1 antibody (2 μg/mL, Santa Cruz Biotechnology, Inc), and incubated for 1 hour with a rabbit anti-mouse IgG conjugated with horseradish peroxidase (dilution 1:50; preabsorbed overnight at 4°C with 10% preimmune rat serum and 3% BSA). Antibody binding in the vessel wall was visualized with 3,3′-diaminobenzidene (Sigma Chemical Co) in 0.1 mol/L Tris buffer (pH 7.2) containing 0.01% H2O2. Sections were counterstained with Harris hematoxylin, dehydrated, and mounted with dePex mounting medium. The number of cells expressing p27kip1 was quantified as the ratio of nuclear-positive cells on total cells in each field.
At the time of the final experiment, the arteries were dissected free from the surrounding tissues, and rats were euthanized for structural and immunohistochemical analyses.7 All data are mean±SEM. ANOVA for repeated measures was performed using the SPSS program, version 10.0; the Tukey test was used to compare single mean values.17
We first determined whether the expression of AKAP75, a prototypic member of the AKAP family, influences cAMP signaling in cultured aortic SMCs, by measuring the transcriptional activity of a cAMP-inducible promoter driving the expression of the bacterial enzyme, CAT (CRE-CAT) (Figure 1⇓).
Expression of AKAP75 induced CRE-CAT in the absence of cAMP stimulation (Figure 1⇑). This effect was mediated by PKA, given that it was reversed by the simultaneous transfection of the PKA nuclear inhibitor PKI, but not by a single-point PKI mutant that did not inhibit PKA (data not shown). These effects were specific to the CRE promoter because RSV-CAT under the same conditions was not activated by AKAP75 or cAMP (Figure 1⇑). Moreover, expression of AKAP45, a derivative of AKAP75 carrying a deletion in the membrane-anchoring domain (residues 1 to 180; Reference 1212 ), did inhibit CRE-CAT both in the presence and the absence of cAMP (Figure 1⇑). Because AKAP45 binds the endogenous PKA and translocates it to the cytosol,12 the inhibition of CRE-CAT transcription is apparently dependent on the loss of cellular membrane-bound PKA. Mutation of the PKA binding site in AKAP45 (A45mut) completely eliminated the negative effect of AKAP45 on cAMP-induced transcription, indicating that both the positive (AKAP75) and the negative (AKAP45) effects on CRE-CAT transcription were mediated by the PKA binding domain of AKAP (Figure 1⇑). This finding also indicates the high degree of specificity of AKAP75 in cAMP signal transduction, because AKAP75 binds other signaling molecules, such as PKC and calcineurin, that might be activated by injury.18
To test the biological consequences of AKAP75 expression, we analyzed the DNA synthesis in AKAP75-transfected cells. Figure 2⇓ shows that the DNA synthesis of AKAP75-expressing cells was dramatically reduced. As control we used cells transfected with AKAP45mut indicated above.
Mock or RSV-NEO–expressing cells show a profile of DNA synthesis and proliferation comparable with those of AKAP45mut-expressing cells (data not shown). Because levels of p27kip1, a specific cyclin-dependent kinase-2 (cdk2) inhibitor, are stimulated by cAMP,19 we determined p27kip1 concentration in AKAP75-expressing cells by Western blotting analysis. Figure 3⇓ shows that cells expressing AKAP75 contained higher p27kip1 levels relative to control cells. Moreover, these cells responded very efficiently to cAMP by increasing p27 levels. PKI but not the mutated variant reduced cAMP-induced p27 levels in AKAP75-expressing cells. These data replicate the effects of AKAP75 on CRE-CAT transcription and indicate that p27 levels are tightly controlled by cAMP-PKA.
To translate these results in vivo, we determined the biological consequences of AKAP75 expression in rat carotid arteries (n=64), subjected to vascular injury as previously described.7 Rat carotid arteries subjected to balloon injury were “transfected” in vivo with plasmid vectors expressing AKAP75, AKAP45, and PKI wild-type or its mutated variant (PKImut). We first determined, 48 hours after the injury, the expression of exogenous AKAP75 gene in the arterial walls with antibodies specific to AKAP75. Cross sections of the arterial walls showed the specific AKAP75 signal in the SMCs of the tunica media of the vessel only in AKAP75-treated animals (Figure 4⇓). The expression was maximal 48 hours after the injury and steadily decreased thereafter (data not shown).
In parallel experiments, the animals were euthanized 14 days after injury to determine the effects of AKAP75 expression on in vivo SMC proliferation. Figure 6⇓ shows that AKAP75 expression reduced significantly the formation of neointima 14 days after arterial balloon injury. DNA expressing vectors containing different viral (RSV-CMV) or eukaryotic (transgenic) promoters driving the expression of bacterial or animal proteins did not inhibit the formation of the neointima.8
The arteries treated with AKAP75 in vivo were also stained with p27 antibody. Figure 5⇓ shows that AKAP75-treated arteries contained a significant fraction of cells that expressed p27kip1 48 hours after the arterial injury (33% versus 18% of positive cells per field, P<0.05) and were not present in arteries transfected with control vectors.
Figure 6⇓ shows representative cross sections of each group of animals studied. The inhibition of neointimal hyperplasia after balloon injury produced by AKAP75 transfection was totally abolished by the wild-type PKA-specific inhibitor PKI14 but not by its mutated version. The neointima area was 0.129±0.040 mm2 in the group transfected with AKAP75 and PKI (n=5) compared with 0.079±0.017 mm2 in the group treated with AKAP75 alone (n=12) (P<0.01 versus AKAP75±PKI). Similarly, the neointima/media ratio was 0.621±0.123 in AKAP75 treated arteries compared with 1.118±0.316 in the group treated with AKAP75 and PKI (P<0.01 versus AKAP75). On the other hand, after balloon injury in the group treated with AKAP75 and PKI mutant, no difference was observed either in neointimal area (n=4) (0.073±0.014 mm2; P=NS versus AKAP75) or in neointima/media ratio (0.717±0.142; P=NS versus AKAP75) compared with the AKAP75 group.
The AKAP75 mutant, AKAP45, carrying the deletion of the membrane-anchoring domain, did not inhibit SMC proliferation in vitro (Figure 2⇑) nor the neointimal formation after balloon injury (Figure 6⇑). In the AKAP45-treated group (n=7), at 14 days after balloon injury the neointimal area was 0.129±0.024 mm2 (P<0.01 versus AKAP75). Similarly, the neointima/media ratio was 0.621±0.123 (P<0.01 versus AKAP75). These findings indicate that AKAP75 effects are tightly dependent on the PKA anchoring domain and that amplification of cAMP signaling reduces neointimal hyperplasia after balloon injury.
The major findings of the present study are that AKAP75 expression in cultured SMCs in vitro and in arteries subjected to balloon injury amplifies and stimulates cAMP-PKA signaling resulting in activation of cAMP-induced transcription, increased levels of p27kip1, and suppression both of SMC growth in vitro and of neointimal hyperplasia after balloon injury.
These effects in vitro and in vivo were dependent on selective stimulation of AKAP-targeted PKAII. Indeed, a PKA-specific inhibitor, PKI, but not its inactive mutant, inhibits the effects of AKAP both in vivo and in vitro. AKAP75 binds and anchors the regulatory subunit of PKA to discrete subcellular membranes. Apparently, the membrane system where AKAP-PKA localizes can vary and depends on the cell type.5 20 21 We have shown that independently on the specific localization of the complex AKAP75-PKA in different cell types (kidney, thyroid, and neuronal cells), the invariant result of AKAP75 expression is the amplification of cAMP signaling to the nucleus. PKA in AKAP75-expressing cells appears to be more stable.5 11 12 Although we did not explore the specific localization of the complex AKAP-PKAII in SMCs, the net effect of AKAP75 expression in cultured SMCs in vitro and in arteries in vivo was the amplification of cAMP signaling.
The selective inhibition of AKAP75 effects by selected mutants provides important information about the mechanism(s) underlying the AKAP75 function. A mutant of AKAP75, AKAP45, which lacks the targeting domain, does not prevent neointimal hyperplasia after balloon injury, inhibits CRE-CAT transcription, and decreases p27kip1. This finding indicates that the RII binding domain is essential for the effects of AKAP75 on cAMP signaling. Moreover, it suggests a function for the endogenous membrane-bound PKA in SMCs. Thus, AKAP45 inhibits basal and cAMP-stimulated transcription, whereas it has no effect on the formation of neointima (Figure 1⇑ and Figure 6⇑). Because this mutant binds endogenous PKA and translocates it to the cytosol,11 12 the inhibition of CRE-CAT transcription and the loss of growth-inhibitory effect suggest that in SMCs endogenous membrane-bound PKA amplifies the effects of cAMP signaling on transcription and inhibits proliferation. We suggest that cytosolic PKA bound to AKAP45 is readily dissociated by basal cAMP, and it is degraded faster. Cells expressing AKAP45 do not transcribe efficiently cAMP-induced genes, contain less p27, and grow faster.12 Furthermore, the effects of AKAP45 are critically dependent on the interaction with PKA, given that the derivative mutant in the RII binding site (AKAP45mut) did not inhibit CRE-CAT transcription or SMC growth (Figures 1⇑ and 2⇑). These data further support the notion that the effects of AKAP on neointima formation after balloon injury are mediated by PKA. Interestingly, AKAP75 binds other signaling molecules, such as PKC or calcineurin, which might independently be activated by arterial balloon injury and might contribute to the regulation of neointima formation as well. Our data suggest that the mechanism of SMC growth inhibition in vivo and in vitro by AKAP75 is related to the amplification of cAMP signals, which can inhibit transcription of cyclin D1 expression15 and increase p27 levels. We have data indicating that cAMP-PKA signals influence the degradation and not the synthesis of p27.12 Under these conditions, even a small quantitative effect (≈3-fold) on p27 levels may have a dramatic effect on DNA synthesis, because p27 levels in G1 are in equilibrium with cdk2, and any variation in this ratio can trigger cdk2 activity.12
The present study further supports the hypothesis that ras pathway inhibition and cAMP stimulation are powerful means to inhibit neointimal formation after balloon injury.8 10 In conclusion, expression of AKAP75 in vitro stimulated cAMP-induced transcription, increased the levels of the cdk2 inhibitor p27kip1, and reduced SMC proliferation. In an in vivo model of vascular injury, AKAP75 significantly increased p27kip1 levels and inhibited neointimal hyperplasia. These effects on SMCs in vitro and on injured arteries in vivo were specifically dependent on the amplification of cAMP-PKA signals by membrane-bound PKA. These data indicate that AKAP proteins selectively amplify cAMP-PKA signaling in vitro and in vivo and support a possible target for the inhibition of the neointimal hyperplasia after vascular injury.
This study was supported in part by Associazione Italiana per la Ricerca sul Cancro, Ministero dell’Universitá e della Ricerca Scientifica e Technologica, and Consiglio Nazionale delle Ricerche (targeted project, biotechnology). We thank Dr CS Rubin (Albert Einstein College of Medicine, Bronx, NY) for kindly providing anti-AKAP75 antibody.
Original received January 18, 2000; resubmission received August 10, 2000; revised resubmission received December 15, 2000; accepted December 15, 2000.
↵1 Both authors contributed equally to this study.
- © 2001 American Heart Association, Inc.
Grieco D, Porcellini A, Avvedimento VE, Gottesman ME. Requirement for cAMP-PKA pathway activation by M phase-promoting factor in the transition from mitosis to interphase. Science. 1996;271:1718–1722.
Brandon EP, Idzerda RL, McKnight GS. PKA isoforms, neural pathways, and behaviour: making the connection. Curr Opin Neurobiol. 1997;3:397–403.
Cassano S, Gallo A, Buccigrossi V, Porcellini A, Cerillo R, Gottesman ME, Avvedimento EV. Membrane localization of cAMP-dependent protein kinase amplifies cAMP signaling to the nucleus in PC12 cells. J Biol Chem. 1996;271:29870–29875.
Indolfi C, Esposito G, Rapacciuolo A, Di Lorenzo E, Feliciello A, Porcellini A, Avvedimento VE, Condorelli M, Chiariello M. Smooth muscle cell proliferation is proportional to the degree of balloon injury in a rat model of angioplasty. Circulation. 1995;92:1230–1235.
Cassano S, Di Lieto A, Cerillo R, Avvedimento EV. Membrane-bound cAMP-dependent protein kinase controls cAMP-induced differentiation in PC12 cells. J Biol Chem. 1999;274:32574–32579.
Feliciello A, Gallo A, Mele E, Porcellini A, Troncone G, Garbi C, Gottesman ME, Avvedimento EV. The localization and activity of cAMP-dependent protein kinase affect cell cycle progression in thyroid cells. J Biol Chem. 2000;275:303–311.
Sassone-Corsi P, Visvader J, Ferland L, Mellon PL, Verma IM. Induction of proto-oncogene fos transcription through the adenylate cyclase pathway: characterization of a cAMP-responsive element. Genes Dev. 1988;2:1529–1538.
Day RN, Walder JA, Maurer RA. A protein kinase inhibitor gene reduces both basal and multihormone-stimulated prolactin gene transcription. J Biol Chem. 1989;264:431–436.
Warnke R, Levy R. Detection of T and B cell antigens hybridoma monoclonal antibodies: a biotin-avidin-horseradish peroxidase method. J Histochem Cytochem. 1980;28:771–776.
Dixon WJ, Massey J. Introduction to Statistical Analysis. New York: McGraw-Hill; 1969.
Skaletz-Rorowski A, Waltenberger J, Muller JG, Pawlus E, Pinkernell K, Breithardt G. Protein kinase C mediates basic fibroblast growth factor-induced proliferation through mitogen-activated protein kinase in coronary smooth muscle cells. Arterioscler Thromb Vasc Biol. 1999;19:1608–1614.
Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev. 1995;9:1149–1163.
Paolillo M, Feliciello A, Porcellini A, Garbi C, Bifulco M, Schinelli S, Ventra C, Stabile E, Ricciarelli G, Schettini G, Avvedimento VE. The type and localization of cAMP signals to the nucleus in cortical and granule cerebellar cells. J Biol Chem. 1999;274:6546–6552.
Feliciello A, Giuliano P, Porcellini A, Garbi C, Obici S, Mele E, Angotti E, Grieco D, Amabile G, Cassano S, Li Y, Musti AM, Rubin CS, Gottesman ME, Avvedimento EV. The v-Ki-Ras oncogene alters cAMP nuclear signaling by regulating the location and the expression of cAMP-dependent protein kinase IIβ. J Biol Chem. 1996;271:25350–25359.