N-Cadherin Upregulation and Function in Response of Smooth Muscle Cells to Arterial Injury
Objective— Smooth muscle cell migration is critical to neointimal formation after arterial injury. The purpose of this study was to elucidate the regulation and functional significance of cell-cell adhesion via adherens junctions during this process.
Methods and Results— Using balloon catheter injury of rat carotid artery, we showed that neointimal formation is accompanied by dramatic but transient upregulation of intimal N-cadherin and associated catenins, proteins that mediate adhesion at adherens junctions. Upregulation was demonstrated by immunofluorescence microscopy and by immunoblotting, and it coincided with evidence of phenotypic modulation of smooth muscle cells. Similar upregulation was observed when postconfluent cultures of porcine aortic smooth muscle cells were subjected to linear denuding injuries. Furthermore, treatment of wounded cultures with a blocking antibody against the extracellular domain of the N-cadherin protein significantly suppressed the repair of wounds.
Conclusions— N-cadherin and associated proteins are dynamically regulated during neointimal formation and provide evidence that this regulation is important for migratory repair. Therefore, N-cadherin may provide a novel target for therapies that are directed toward intimal proliferative disorders, including restenosis and vascular bypass graft failure.
Smooth muscle cell migration and proliferation contribute to the intimal thickening that characterizes atherosclerosis, failure of vascular bypass grafts, and restenosis. Migration is accompanied by a transition from a contractile quiescent state to a proliferating synthetic phenotype. Concomitantly, cell-matrix adhesion is dynamically regulated by the coordination of adhesion and release events that enable cell movement.1–5⇓⇓⇓⇓
Smooth muscle cells also dynamically modulate physical interactions with neighboring cells during migration; however, cell-cell adhesion during neointimal formation has been infrequently studied. Adherens junctions are major cell-cell adhesion complexes that constitute transmembrane cadherin molecules that undergo homophilic binding and that link intracellularly to the actin cytoskeleton via structural and signaling molecules, the catenins. Vascular smooth muscle cells express multiple cadherins, including N-cadherin, R-cadherin, T-cadherin, cadherin-6b, and E-cadherin (in atherosclerotic lesions).6 We focused on N-cadherin, the primary cadherin expressed by vascular smooth muscle cells, because of the novel role that this cadherin plays in cell movement.7–10⇓⇓⇓ N-cadherin, like other cadherins, can mediate contact inhibition of cell growth11 and migration12; however, N-cadherin expression is increased in migrating cells undergoing epithelial-mesenchymal transformation during myocardial morphogenesis,13 when epiblast cells ingress through the primitive streak to form the mesoderm in the developing chick embryo,14 and in metastatic squamous epithelial cell lines.15–17⇓⇓ Furthermore, N-cadherin antibodies inhibit neurite outgrowth.18,19⇓ These studies indicate that N-cadherin–based adherens junctions are dynamic structures that can mediate transient cell-cell adhesions during tissue remodeling. Finally, our initial findings indicated that N-cadherin is upregulated after arterial injury, whereas R-cadherin, T-cadherin, and cadherin-6b are all downregulated.6
We have observed upregulation of N-cadherin and catenins during neointimal formation after injury of the rat carotid artery and after wounding of smooth muscle cell cultures. Furthermore, antibodies that blocked homophilic binding of N-cadherin suppressed spreading and migration during repair of in vitro wounds. Therefore, N-cadherin may provide a novel target for therapeutic interventions that seek to limit intimal smooth muscle cell accumulation in vascular disorders, including restenosis and vascular bypass graft failure.
An extended Methods section is provided in an online supplement (please see www.ahajournals.org).
Balloon catheter injury of the carotid artery in male Sprague-Dawley rats weighing 350 to 400 g was performed with the animals under general anesthesia. The animals were allowed to recover for 1, 3, or 8 weeks.
In Vitro Wound Migration Assay
Postconfluent cultures of porcine aortic smooth muscle cells (passages 4 to 9) were injured by drawing a pipette tip across the cells to create a wound that was ≈800 μm wide. Wound widths were measured by phase-contrast microscopy during closure. Mean values of percent closure of wounds were subjected to 1-way ANOVA, followed by Dunnett tests (n=5).
For biochemical assays, wounds were made by drawing a comb device across the cultures twice, in perpendicular directions. The 1-mm-wide teeth of the comb were separated by 2 mm.
Control and injured carotid arteries were harvested and cleared of adventitia, and proteins were extracted for Western blotting. Proteins were also extracted from cultured porcine aortic smooth muscle cells that were lysed in buffer. Western blots were performed by using a pan-cadherin antibody or antibodies against N-cadherin, β-catenin, or plakoglobin. Bands were visualized with an enhanced chemiluminescence Western blotting detection system.
Band volume intensity was quantified by densitometry (Quantity One Software, Version 4.2.3, Bio-Rad Laboratories). All comparisons were made relative to control conditions.
Rats were killed and fixed by perfusion at a pressure of 100 mm Hg. Routine histological sections were immunostained by using primary antibodies that recognized N-cadherin, β-catenin, plakoglobin, or smooth muscle α-actin antibody that recognizes classic cadherins. F-actin was stained with rhodamine phalloidin. Sections were viewed with a Bio-Rad confocal microscope. Alternatively, uninjured and injured carotids were perfusion-fixed and incubated en face with the antibodies, and then whole-mount preparations, with the intimal side up, were viewed by confocal microscopy.
Cultured porcine aortic smooth muscle cells were fixed with 3% paraformaldehyde, permeabilized with 0.2% Triton X-100 for 5 minutes, rinsed again with PBS, and incubated with primary antibodies as described above. Alternatively, unpermeabilized cells were stained with propidium iodide to label cell nuclei and 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI) to label the plasma membrane. The cells were viewed by confocal microscopy as described above.
Cell spreading into wounds was assessed by measuring the distance, perpendicular to the wound, from the leading edge of the propidium iodide–labeled cell nuclei to the leading edge of the cell, which was detected by DiI staining.
At defined times after wounding, 20 μL of conditioned media from each well was loaded on an SDS-PAGE gel containing 0.1% gelatin as a substrate for matrix metalloproteinase (MMP) activity. Zymograms were processed and stained as we have previously described.20
Phenotypic Modulation and Upregulation of Adherens Junction Proteins During Neointimal Formation
As previously described,21 balloon catheter injury to rat carotids resulted in the formation of a neointima that first appeared after 4 days and then increased in thickness for at least 8 weeks. Phenotypic modulation of intimal smooth muscle cells was evident. Medial smooth muscle cells were fusiform and oriented circumferentially, and the cytoplasm was rich with F-actin filaments (Figure 1A). Intimal cells were rounder and randomly oriented, and total F-actin was sparser and concentrated at the cell periphery (Figure 1B). Furthermore, immunostaining with an antibody specific for smooth muscle α-actin revealed high levels in the media of control and injured arteries (Figure 1C) but an almost complete absence of this actin isoform in the neointima at 1 week (Figure 1C).5 At 3 and 8 weeks, smooth muscle α-actin was localized close to the plasma membrane of neointimal cells (Figure 1D).
Intimal levels of N-cadherin, β-catenin, and plakoglobin were upregulated during neointimal formation. At 1 and 3 weeks after injury, neointimal immunostaining for N-cadherin was consistently more intense than that of the media, and then it fell to medial levels at 8 weeks after injury (Figure 2B through 2D). Medial staining and adventitial staining were comparable across all the sections and comparable to the staining of uninjured contralateral control arteries (Figure 2A). Minimal staining was observed when sections were stained with secondary antibody only or when anti-cadherin antibody was preincubated with the peptide against which it was raised (not shown). Similar results were observed with immunostaining for β-catenin and plakoglobin (please see online Figures I and II, available at www.ahajournals.org). Viewing the vessels en face with the use of confocal microscopy revealed that adherens junction proteins had localized to the cell periphery (please see online Figure III, available at www.ahajournals.org); furthermore, serial optical sections confirmed that the proteins were distributed around the entire plasma membrane of the cell (data not shown).
Immunoblots of carotid lysates confirmed the inferences drawn from immunofluorescence staining (Figure 2E). Blots displayed a single band at 135 kDa when probed for N-cadherin, a single band at 92 kDa when probed for β-catenin, and a doublet at 82 kDa when probed for plakoglobin. The plakoglobin doublet may represent phosphorylated and dephosphorylated forms of the protein. Levels of all 3 proteins were maximally elevated at 1 week after injury. Densitometry revealed significant (P<0.05) elevations of N-cadherin (2.9-fold), β-catenin (1.6-fold), and plakoglobin (2-fold) at this time. Protein levels declined to those of uninjured carotids by 8 weeks.
Adherens Junctions in Cultured Vascular Smooth Muscle Cells
We exploited a standard cell culture wound model to test whether upregulation of adherens junction proteins during neointimal formation is important in smooth muscle cell responses to injury. Because little research has been performed on adherens junctions in vascular smooth muscle cells, we first characterized the junctions in this cell type.
When immunoblots of smooth muscle cell lysates were probed with a pan-cadherin antibody, a single band was detected at 135 kD, and immunoblotting with the use of a monoclonal anti–N-cadherin antibody confirmed that this protein was N-cadherin (data not shown). No band corresponding to E-cadherin was detected. Immunostaining revealed that N-cadherin was localized at the tips of the cell projections between contacting cells, at the ends of F-actin microfilament bundles (Figure 3A). Immunofluorescence microscopy confirmed that α-catenin (Figure 3B), β-catenin (Figure 3C), and plakoglobin (Figure 3D) were also localized at these sites. Confocal optical sections collected at a separation of 0.25 μm demonstrated that the adherens junctions were localized at the midregion of the cell (data not shown). Nonclassic cadherins (T- or R-cadherin or cadherin-6b) were not examined.
N-Cadherin–Mediated Cell-Cell Adhesion Promotes Migration of Smooth Muscle Cells After Wounding In Vitro
After scratch wounding of smooth muscle cell cultures, immunostaining patterns for adherens junction proteins were similar to those of intact monolayers, except that the proteins were excluded from free edge of cells that extended into the wound (please see online Figure IVA, available at www.ahajournals.org). In some other cell systems, plakoglobin is restricted to stable long-term adherens junctions22,23⇓; however, immunostaining revealed no obvious shift in plakoglobin versus β-catenin that was associated with adherens plaques. As in arterial injury in vivo, increased protein levels were associated with wound repair. Immunoblots from extracts of cell cultures that were comb-wounded revealed upregulation of N-cadherin (please see online Figure IVB), β-catenin, and plakoglobin (not shown).
We next determined whether a blocking antibody to N-cadherin would alter cell migration and wound closure. At the wound edge, control smooth muscle cells or cells incubated with nonspecific mouse IgG displayed long thin processes, and the cells rapidly migrated into the wound. Incubation of cultured smooth muscle cells with anti–N-cadherin antibody during wound repair caused a significant reduction in the rate of wound repair. Statistical analysis of percent wound closure against time revealed a significant, ≈50%, slowing of wound repair in antibody-treated cultures at 24 hours and 48 hours (P<0.05, Figure 4).
Repair at 24 and 48 hours may be dominated by cell replication. To test for specific effects on cell spreading and migration into the wound site, we examined repair at 12 hours after wounding, before the cells induced to replicate by the injury had entered the S phase of the cell cycle. At this time, as at later times, cells in control cultures displayed a high degree of orientation toward the wound (Figure 5A), whereas cells in cultures treated with anti–N-cadherin antibody did not display well-defined orientation (Figure 5B). Furthermore, protrusion of the cells on the wound side of the nucleus was dramatically suppressed by antibody treatment (P<0.005, Figure 5C). This suppression was due to decreased spreading and also to misdirected movement of the cells into the wounds. These findings indicate that spreading and/or migration into the wound is sensitive to N-cadherin–mediated cell adhesion. Impaired spreading was not due to effects on MMP production. Zymography revealed that wounded cultures released latent and active MMP-2 and active MMP-9 into the media, but this release was unaffected by anti–N-cadherin antibody (data not shown).
In the present study, intimal proliferation after balloon catheter injury proceeded as described previously.21 Intimal smooth muscle cells were readily detectable at 1 week after injury, and thickening of the neointima continued for at least 8 weeks. Injury was followed by phenotypic modulation of smooth muscle cells in the rat carotid artery, and confocal microscopy provided a novel perspective on aspects of this phenomenon. Reorganization of F-actin in smooth muscle of the early neointima, with concentration predominantly at the cell periphery, was readily apparent. Furthermore, the complete absence of smooth muscle α-actin in the early intima, as shown by Pickering et al,5 was confirmed.
Although smooth muscle cell–matrix interactions have been extensively examined in the context of the forming neointima, cell-cell adhesion complexes have not been studied. Adherens junctions are primary sites of mechanical adhesion between cells that may be critical to spreading, migration, and local tissue remodeling after mitosis. N-cadherin, thought to be the primary cadherin expressed by vascular smooth muscle cells, is also expressed by fibroblasts, cardiac myocytes, skeletal muscle, lens epithelial cells, and vascular endothelium.7–10,24⇓⇓⇓⇓ Adherens junctions, including those that harbor N-cadherin, can mediate contact inhibition of cell growth11 and migration12; however, the role of N-cadherin in mesenchymal cells is more complex. During myocardial morphogenesis, N-cadherin expression is significantly increased in cells undergoing epithelial-mesenchymal transformation and migrating toward the endocardium during the formation of the cardiac trabeculae and intercalated disks.13 Similarly, in the developing chick embryo, there is a switch from E- to N-cadherin when epiblast cells ingress through the primitive streak to form the mesoderm.14 In addition, N-cadherin expression is upregulated in some squamous epithelial cell lines that have become metastatic15,16⇓ despite continuous expression of endogenous E- and P-cadherins.17 The latter finding suggests that N-cadherin overrides the ability of E-cadherin to maintain an epithelial phenotype and stable junctions in some cells. This may be particularly relevant inasmuch as E-cadherin is expressed by smooth muscle in atherosclerotic lesions. Finally, the involvement of N-cadherin in cell motility is also demonstrated by the capacity of N-cadherin antibodies to inhibit neurite outgrowth.18,19⇓
N-cadherin and associated catenins in smooth muscle cells were upregulated during neointimal formation, a finding that implicates these junctions in the cellular response to arterial injury. Furthermore, in smooth muscle cell cultures, N-cadherin participated in the formation of junctional complexes that localized to puncta at cell-cell interfaces. Although this distribution was more restricted than that seen in vivo, similar alterations to adherens junction protein contents occurred after in vivo and in vitro injury. N-cadherin and the catenins were again upregulated during repair and returned to control levels after reestablishment of confluence in vitro, just as the proteins were upregulated while neointimal formation progressed in vivo. These findings are consistent with a dynamic role for these junctions in smooth muscle cell responses to injury.
Inhibition of N-cadherin–mediated adhesion with a function-blocking antibody suppressed in vitro wound repair after denuding the wounds. Inhibition was observed at 24 and 48 hours; therefore, either migration or early cell proliferation could be involved. Indeed, a recent study strongly implicated β-catenin signaling in the regulation of cell proliferation after arterial injury.25 However, we noted that early spreading and migration of cells into the wound site appeared to be inhibited by the antibody. Subsequent studies focused on repair in the first 12 hours after wounding, before cells enter the S phase of the cell cycle. We found that movement into the wound site in control cultures was a highly oriented process with leading edge cells aligning perpendicularly to the wound. Movement became highly disoriented after antibody treatment; furthermore, protrusion of the leading edge of the cell, proximal to the nucleus, was substantially and significantly suppressed. These findings suggest that homophilic binding of N-cadherin only at the trailing edge of cells at the wound margin may provide cues in defining the direction of repair.
N-cadherin may also promote wound repair by forming labile cellular adhesions that facilitate dynamic cell-cell interactions.26,27⇓ Labile contacts would allow the release at some adherens complexes while propulsive forces are being generated at others. However, the mutual motility of cells that are coupled by N-cadherin is complex and related to the viscoelastic properties of the intracellular milieu,28 so novel mechanisms of migration may be elicited. Regardless of the mechanisms of cell movement, blockade of adhesion mediated by N-cadherin complexes may limit labile interactions so that smooth muscle cells can no longer navigate past one another.
Plakoglobin and β-catenin compete for a binding site on the cytoplasmic domain of cadherins to link these transmembrane proteins to the actin cytoskeleton via α-catenin. In many cell types, plakoglobin localizes only to very stable long-term adherens junctions; in contrast, β-catenin can localize to junctional complexes while these structures are reorganizing. For example, in vascular endothelium, plakoglobin is the first protein to dissociate from adherens junctions during the repair of denuding wounds as cells spread, elongate, and migrate.23 Similarly, shear stress–induced endothelial cell shape change is accompanied by dissociation of plakoglobin, but not β-catenin, from cell-cell junctions until a new steady-state cell morphology is established.29 These findings suggested that a transition from plakoglobin-based to β-catenin–based adhesion might underlie the transition to a more labile junction. However, our observation, ie, that increases in plakoglobin protein levels and adherens junction localization matched those of β-catenin and N-cadherin during wound repair, does not support this interpretation.
The adherens junction may also modulate migration through intracellular interactions with integrin-based adhesion complexes. For example, under the influence of proteoglycan-proteoglycan receptor interactions, the nonreceptor tyrosine kinase, Fer, can be transferred from the cytoplasmic domain of N-cadherin to β1 integrins, a phenomenon that suppresses neurite outgrowth from retinal explants by suppressing integrin- and cadherin-mediated adhesion.30,31⇓ In our experiments, antibody treatment of wounded smooth muscle cell cultures suppressed the spreading of cells into wound areas; furthermore, very high concentrations of antibody induced rounding of cells suggestive of compromised cell-substrate adhesions (data not shown). These findings are consistent with the hypothesis that disrupting cadherin function inhibits integrin-matrix interaction in smooth muscle.
Finally, N-cadherin may interact with growth factor signaling that promotes migration. Homophilic binding of N-cadherin between neuronal growth cones and astrocytes, oligodendrocytes, and Schwann cells mediates neurite outgrowth.32 In this setting, N-cadherin interacts with and activates the fibroblast growth factor (FGF) receptor through dimerization. FGF receptor activation through N-cadherin may also play a role in the ability of cancer cells of epithelial origin to infiltrate host tissue. FGF-2 stimulates cancer cell migration and production of MMP-2 but only in those cells expressing N-cadherin. This finding suggests that cell adhesion molecules, growth factor–mediated signals, and proteolysis of the extracellular matrix are involved in the concerted action of cell migration, invasion, and metastasis.33
In summary, we have shown that N-cadherin and catenins are dynamically regulated in smooth muscle cells during migratory responses to arterial injury. We have also demonstrated that inhibition of N-cadherin function suppresses smooth muscle migration during repair. These findings elucidate novel mechanisms by which arterial responses to injury are mediated, and they suggest that N-cadherin–mediated adhesion may provide a novel therapeutic target for treatment of vascular disorders that elicit intimal proliferation, including restenosis and bypass graft failure.
This work was supported by the Canadian Institutes for Health Research (grant 15044).
Received July 15, 2002; revision accepted August 15, 2002.
- ↵Bendeck MP, Irvin C, Reidy MA. Inhibition of matrix metalloproteinase activity inhibits smooth muscle cell migration but not neointimal thickening after arterial injury. Circ Res. 1996; 78: 38–43.
- ↵Hoshiga M, Alpers CE, Smith LL, Giachelli CM, Schwartz SM. αvβ3 integrin expression in normal and atherosclerotic artery. Circ Res. 1995; 77: 1129–1135.
- ↵Moiseeva EP. Adhesion receptors of vascular smooth muscle cells and their functions. Cardiovasc Res. 2001; 52: 372–386.
- ↵Knudsen KA, Soler AP, Johnson KR, Wheelock MJ. Interaction of α-actinin with the cadherin/catenin cell-cell adhesion complex via α-catenin. J Cell Biol. 1995; 130: 67–77.
- ↵Huttenlocher A, Lakonishok M, Kinder M, Wu S, Truong T, Knudsen KA, Horwitz AF. Integrin and cadherin synergy regulates contact inhibition of migration and motile activity. J Cell Biol. 1998; 141: 515–526.
- ↵Edelman GR, Gallin WJ, Delouvee A, Cunningham BA, Thiery JP. Early epochal maps of two different cell adhesion molecules. Proc Natl Acad Sci U S A. 1983; 80: 4384–4388.
- ↵Li Z, Gallin WJ, Lauzon G, Pasdar M. L-CAM expression induces fibroblast-epidermoid transition in squamous carcinoma cells and down-regulates the endogenous N-cadherin. J Cell Sci. 1998; 111: 1005–1019.
- ↵Islam S, Carey TE, Wolf GT, Wheelock MJ, Johnson KR. Expression of N-cadherin by human squamous carcinoma cells induces scattered fibroblastic phenotype with disrupted cell-cell adhesion. J Cell Biol. 1996; 135: 1643–1654.
- ↵Nieman MT, Prudoff RS, Johnson KR, Wheelock MJ. N-cadherin promotes motility in human breast cancer cells regardless of their E-cadherin expression. J Cell Biol. 1999; 147: 631–643.
- ↵Bixby JL, Zhang R. Purified N-cadherin is a potent substrate for the rapid induction of neurite outgrowth. J Cell Biol. 1990; 110: 1253–1260.
- ↵Bendeck MP, Zempo M, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res. 1994; 75: 539–545.
- ↵Noria SF, Jackson MR, Langille BL. Vascular adaptations to altered blood flow. In: Colman RW, Hirsh J, Marder VJ, Clowes AW, George JN, eds. Hemostasis and Thrombosis: Basic Principles & Clinical Practice. Philadelphia, Pa: Lippincott Williams & Wilkins; 2001: 733–742.
- ↵Lampugnani MG, Corada M, Caveda L, Breviario F, Ayalon O, Geiger B, Dejana E. The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, β-catenin, and α-catenin with vascular endothelial cadherin (VE-cadherin). J Cell Biol. 1995; 129: 203–217.
- ↵Salomon D, Ayalon O, Patel-King R, Hynes RO, Geiger B. Extrajunctional distribution of N-cadherin in cultured human endothelial cells. J Cell Sci. 1992; 102: 7–17.
- ↵Wang X, Xiao Y, Mou Y, Zhao Y, Blankesteijn WM, Hall JL. A role for the β-catenin/T-cell factor signaling cascade in vascular remodeling. Circ Res. 2002; 90: 340–347.
- ↵Noria S, Cowan DB, Gotlieb AI, Langille BL. Transient and steady-state effects of shear stress on endothelial cell adherens junctions. Circ Res. 1999; 85: 504–514.
- ↵Arregui C, Pathre P, Lilien J, Balsamo J. The nonreceptor tyrosine kinase fer mediates cross-talk between N-cadherin and beta1-integrins. J Cell Biol. 2000; 149: 1263–1274.
- ↵Li H, Leung TC, Hoffman S, Balsamo J, Lilien J. Coordinate regulation of cadherin and integrin function by the chondroitin sulfate proteoglycan neurocan. J Cell Biol. 2000; 149: 1275–1288.
- ↵Hazan RB, Phillips GR, Qiao RF, Norton L, Aaronson SA. Exogenous expression of N-cadherin in breast cancer cells induces cell migration, invasion, and metastasis. J Cell Biol. 2000; 148: 779–790.