Homotypic and Endothelial Cell Adhesions via N-Cadherin Determine Polarity and Regulate Migration of Vascular Smooth Muscle Cells
Migration of smooth muscle cells from the arterial media to the intima is central to several vascular pathologies including restenosis. This study demonstrates that, like directional migration of other cells, smooth muscle migration is accompanied by a dramatic, polarized reorganization of the cell cytoskeleton that is accompanied by activation of the Rho GTPase Cdc42 and inactivation of glycogen synthase kinase-3β. We also show, for the first time, that signals generated at the posterior–lateral aspects of wound edge cells by the cell–cell adhesion molecule N-cadherin are required for polarization and rapid migration of vascular smooth muscle. Importantly, when a cohort of migrating smooth muscle cells encounter CHO cells or the A10 smooth muscle cell line, neither of which expresses N-cadherin, polarity is only slightly suppressed. However, when smooth muscle cells encounter stably transfected, N-cadherin–expressing A10 cells or (N-cadherin–expressing) vascular endothelium, they rapidly lose their polarized phenotype. The latter finding indicates that endothelial signaling to innermost smooth muscle cells via N-cadherin may be critical to normal vessel wall stability. We infer that asymmetrical distribution of N-cadherin is necessary for the establishment of cell polarity during migration and that N-cadherin ligation is highly effective in abrogating polarized migration. Finally, we showed that endothelial cell polarity does not depend on N-cadherin; therefore, this molecule may be an attractive target for therapies to prevent restenosis without suppressing endothelial repair and risking late thrombosis.
Directional migration of smooth muscle cells from the arterial media to the intima and their subsequent proliferation is critical to the pathogenesis of important vascular disorders including restenosis, atherosclerosis, and bypass graft failure.1,2 Endothelial cell loss initiates this neointimal formation because it occurs, albeit more slowly, after very gentle procedures that achieve deendothelialization with minimal medial trauma,3 whereas severe medial trauma that leaves the endothelium largely intact does not produce a neointima.4 Consequently, the sequelae to endothelial loss have been extensively studied.
Platelet activation after endothelial denudation and subsequent release of the chemotactic mitogen, platelet-derived growth factor (PDGF), are important in promoting inward migration of smooth muscle cells.5 Accordingly, antiplatelet or anti-PDGF strategies slow neointimal formation in animal models; however, migration is not fully arrested, and these strategies have not proven to be of substantial clinical value.6 Instead, recent improvements in restenosis outcomes that have resulted from use of drug-eluting stents for the treatment of coronary artery disease rely on release of antimitogenic agents.7,8 A limitation to antimitogenic therapy is that it may contribute to late thrombosis after stenting by suppressing endothelial repair.9
Investigations of diverse cell types indicate that directional migration is often a cell-autonomous response to loss of neighboring cells. Cultured cells at the edge of experimental wounds migrate toward the wound and display a highly polarized reorganization of cytoskeletal elements that promotes this directional migration.10 Most strikingly, the microtubule system, and the microtubule organizing center (MTOC) from which the system emanates, moves anterior to the cell nucleus, a positioning of microtubules that promotes multiple functions, including transport of cell membrane and structural proteins to and from the leading edge of the cell. Polarized distribution is accompanied by posttranslational modifications to tubulin that hyperstabilize most microtubules, which further enhances cargo transport.11 This type of cell polarity was first described in vascular endothelium12 and has since been studied extensively in other cell types.10,13,14 It depends on the establishment of an intracellular, posterior–anterior gradient in signaling that involves activation of the Rho GTPase Cdc42 and then inhibition of glycogen synthase kinase (GSK)-3β by noncanonical Wnt signaling.15,16
Elegant work with astrocytes and fibroblasts has proven that integrin activation at the cell anterior can initiate the signaling gradient that drives cell polarity and directional migration.17 However, cell–cell contact localizes to the posterior–lateral aspects of wound edge cells; therefore, the intracellular signals that result from this contact are also potential initiators of cell polarity. In this context, we were intrigued by observations that N-cadherin, an important mediator of smooth muscle cell–cell adhesion, suppresses Cdc42 activity.18 Therefore, disinhibition of this signaling at the leading edge of wound edge cells, attributable to loss of cell–cell contact, could elicit the intracellular gradients that drive cell polarity and directional migration. Importantly, N-cadherin mediates adhesion between smooth muscle and endothelial cells19; therefore, loss of endothelium in vivo might elicit such polarized signaling in innermost smooth muscle cells within the media.
In the present study, in vitro wound models were used to demonstrate that N-cadherin is an obligatory source of signaling that drives polarity and accelerates migration of vascular smooth muscle cells and that the downstream signaling acts via regulation of Cdc42 and GSK-3β. Furthermore, we used a model of heterotypic cell contact to demonstrate that when smooth muscle cells contact other cells expressing N-cadherin (endothelial cells or A10 cells transfected with N-cadherin), polarization is dramatically suppressed. By contrast, contact of smooth muscle cells with cells that do not express N-cadherin (CHO or A10 cells) only modestly suppresses polarization. The inhibitory effects of endothelium on smooth muscle polarity and migration may indicate an important role for endothelial signaling to innermost smooth muscle cells via N-cadherin in stabilizing the structure of mature, quiescent arteries. Finally, polarity of migrating endothelium did not depend on N-cadherin–related signals; therefore, N-cadherin provides a useful target for restenosis therapy that might avoid the risk of persistent endothelial denudation and late thrombosis that characterizes antimitogenic therapies.
Materials and Methods
Detailed methods are provided in the online data supplement at http://circres.ahajournals.org.
Porcine arterial smooth muscle cells were used at passages 4 to 9 and porcine endothelial cells at passages 4 to 7. Additional experiments used the MOVAS smooth muscle cell line.20 CHO cells and A10 cells were obtained from the American Type Culture Collection.
Postconfluent cultures were wounded by dragging a 200-μL pipette tip across the bottom of the plate. When indicated, taxol, nocodazole, EGTA, LiCl, or SB415286 (GSK-3β inhibitor) were added to the media 1 hour before wounding.
In some experiments, a monoclonal anti–N-cadherin antibody (GC-4, which binds to the extracellular domain of N-cadherin), or a nonspecific mouse IgG1κ antibody was added to the media 16 hours before wounding to prevent N-cadherin–mediated adhesion. Alternatively, various concentrations of a linear N-cadherin–specific blocking peptide21 (N-Ac-LRAHAVDING-NH2) or a scrambled control peptide (N-Ac-HLNARGAIVD-NH2) was added 1 hour before wounding.
In d and f of Figure 6, the green channel was suppressed within Adobe Photoshop to improve visibility of MTOC (red channel). No other image processing was used.
Quantification of Cell Polarity
Wound edge cells were described as “polarized” if the MTOC (identified using γ-tubulin antibody) was anterior to the midpoint of the nucleus, or “highly polarized” if their MTOC was anterior to the nucleus (Figure 1c). With this scheme, 50% of cells are labeled polarized in randomly oriented cultures. ANOVA followed by Tukey or Bonferroni comparisons were used to establish significance among different time points and treatments, a Student t test was used to compare a single treated group with control cultures, and a Dunnett’s test was used when multiple experimental groups were compared with a single control.
Subconfluent smooth muscle cells were transfected with 4 μg/mL of the Cdc42-GFP plasmid (gift from G. Downey, University of Toronto, Toronto, Canada) with Effectene (Qiagen) transfection kits. To generate stably expressing N-cadherin–GFP or α-tubulin–GFP A10 cell lines, the N-cadherin–GFP (gift from Cecile Gauthier-Rouviere, Centre de Recherches de Biochimie Macromoléculaire, Centre National de la Recherche Scientifique, IFR 122, 34293 Montpellier, France) or α-tubulin-GFP (Clontech) plasmids were transfected into A10 cells with Effectene (Qiagen), and clones resistant to G418 were isolated.
Rho GTPase Activity Assay
Smooth muscle cells were grown to confluence on 90-mm2 Petri dishes and then wounded twice in orthogonal directions with a comb containing 13 protrusions for a total of 26 wounds in a grid-like pattern. Cdc42/Rac1 and RhoA activation assay kits were purchased from Cytoskeleton (Denver, CO) and used to detect active Cdc42, Rac1, and RhoA according to the instructions of the manufacturer.
Cells were grid-wounded as described above and were probed using antibodies against GSK-3, phospho(Ser9)-GSK-3β, Cdc42, Rac1, RhoA, N-cadherin, and horseradish peroxidase–conjugated secondary antibodies. An ECL detection kit (Amersham Bioscience) was used to detect protein, which was quantified by densitometry.
Heterotypic Adhesion and Smooth Muscle Cell Polarity
Cocultures of porcine smooth muscle cells with CHO cells, porcine aortic endothelial cells, α-tubulin–expressing A10 cells (no N-cadherin expression), or N-cadherin–GFP expressing A10 cells were used to assess the effect of differential N-cadherin expression on cell polarity on wound closure. A cloning ring was placed on a glass cover slip, and porcine smooth muscle cells were seeded on the inside at the same time as either endothelial cells, CHO cells, or A10 cells were seeded on the outside of the cloning ring. One day later, the cloning ring was removed to allow heterotypic cell wound repair. At 6 hours after first heterotypic cell contact, the cells were fixed and immunostained for γ-tubulin to identify the MTOC and with cell type–specific antibodies. Nuclei were counterstained with TOTO-3 and the GFP signal was used to differentiate A10 cells from porcine arterial smooth muscle cells.
In Vivo Studies
Animal experiments were performed according to the guidelines of the Canada Council on Animal Care. Adult male Sprague–Dawley rats (Charles River, Constant, Quebec, Canada) weighing 375 to 415 g were anesthetized by IP injection of 4.6 mg/kg xylazine (Rompum; Bayer Inc, Etobicoke, Ontario, Canada) and 70 mg/kg ketamine (Ketaset; Ayerst Veterinarian Laboratories, Guelph, Ontario, Canada). Endothelium was removed from the left carotid artery of adult male Sprague–Dawley rats by passing a loop of 5-0 monofilament nylon into the vessel via the external carotid and withdrawing the nylon while rotating,22 and then incisions were closed. Four days later, arteries were excised, stained with the DNA label propidium iodide and viewed en face.
Migrating Smooth Muscle Cells Display Microtubule-Dependent Cell Polarity
A hallmark of cell polarity of migrating cells is the positioning of the MTOC to the cell anterior relative to the nucleus.10,12,13 By 12 hours after wounding postconfluent cultures of porcine smooth muscle cells, most MTOCs of wound edge cells displayed this feature, as indicated by immunofluorescence staining for the MTOC marker, γ-tubulin (Figure 1a). Furthermore, immunofluorescence staining for acetylated tubulin, a marker of hyperstabilized microtubules, revealed the anterior positioning of these filaments (Figure 1b). Quantification of MTOC polarity indicated that the proportion of cells that were polarized rose from 50% at time of wounding (randomly distributed MTOCs) to >90% within 12 hours of wounding; furthermore, the fraction of the cell population that was highly polarized rose from less than 25% to almost 60% within this time (Figure 1c and 1d).
Repositioning of the MTOC required an intact microtubule system because it was abolished in the presence of the microtubule-disrupting agent, nocodazole (Figure 1e). Microtubules are highly dynamic structures; they polymerize rapidly, and they undergo frequent, extensive, or complete depolymerization (catastrophes). The half-life of these filaments is typically 10 minutes.23 Although many microtubules became hyperstabilized (a half-life that is typically 1 hour),24 dynamic instability remained critical for the establishment of cell polarity because repositioning of the MTOC anterior to the nucleus was prevented when cultures were treated with the microtubule stabilizing agent, taxol (Figure 1e).
To confirm that outward migration from in vitro wounds reasonably reflects in vivo behavior, we examined smooth muscle cell populations in the rat neointima soon after wounding. Four days following endothelial denudation of the carotid artery, en face preparations were stained with propidium iodide and visualized by confocal microscopy. Micrographs established that smooth muscle cells could migrate individually through fenestrae in the internal elastic lamina but that they formed clusters of intimal cells that spread outward (Figure I in the online data supplement). Therefore, smooth muscle cells at the periphery of these intimal colonies are subjected to the same polarized signaling from cell–cell adhesions as wound edge cells in vitro.
N-Cadherin–Mediated Adhesion Is Required for Cell Polarity and Efficient Directional Migration of Smooth Muscle Cells
To test whether N-cadherin signaling is required for cell polarity, we treated smooth muscle cell cultures with antibody against the extracellular domain of the protein to prevent homophilic binding.25,26 Immunofluorescence staining confirmed posterior–lateral localization of N-cadherin on wound edge cells, and it revealed loss of most of this junctional N-cadherin in the presence of N-cadherin antibody (Figure 2a and 2b). As previously reported,27 a slowing of wound repair (supplemental Figure II) was observed in the presence of the N-cadherin antibody, and time lapse phase contrast movies demonstrated that wound edge cells rapidly oriented in the direction of migration in untreated cultures, whereas this orientation was lost in the presence of the antibody (Figure 2c and 2d and supplemental Movies 1 and 2).
Impairment of cell alignment and migration was accompanied by a complete loss of cell polarity, as indicated by MTOC position. MTOCs became randomly positioned (Figure 2e), with the percentage of cells that were highly polarized at 6 hours decreasing from >40% in control cultures to <10% in the presence of antibody (Figure 2f). Control IgG was without effect. N-Cadherin dependence of polarity was confirmed when a N-cadherin–blocking peptide was used instead of antibody, whereas the control peptide had no effect (Figure 2g). Finally, cadherin-mediated adhesion is calcium-dependent; therefore, calcium chelation is often used to disrupt/reestablish adherens junctions (calcium switch experiments).28 Accordingly, EGTA treatment of cultures also prevented cell polarity at wound edges, whereas adding back calcium restored cell polarity within 6 hours (Figure 2f). Dependence of polarity on N-cadherin was not restricted to the establishment of polarity during initiation of movement because late addition of the N-cadherin antibody to already polarized cells (6-hour wound) abolished cell polarity within 6 hours (data not shown).
These interventions unequivocally establish that N-cadherin–mediated adhesion is required for the initiation and maintenance of smooth muscle cell polarity. Similar experiments showed that polarity of an immortalized mouse aortic smooth muscle cell line (MOVAS) also depend on N-cadherin–mediated adhesion (data not shown).
N-Cadherin Regulates Cell Polarity by Controlling Cdc42 Activation
Rho GTPases are important regulators of cell migration and polarity10,13; therefore, we used standard pull-down assays to detect active (GTP-bound) RhoA, Rac1, and Cdc42 after comb-wounding porcine smooth muscle cell cultures (Figure 3). RhoA activity was slightly and transiently suppressed after wounding, whereas Rac activity increased gradually, ultimately displaying a persistent >2-fold elevation over control levels. Notably, Cdc42 activity increased rapidly after wounding and was highest during the first 6 hours. Cdc42 activity returned to resting levels between 6 to 12 hours after wounding.
Cdc42 is a proven regulator of cell polarity,14 and it is sensitive to N-cadherin signaling, at least in C2C12 cells.18 Therefore, we tested whether N-cadherin regulates wound-induced Cdc42 activation. Maximal activation at 3 hours after wounding was completely suppressed when cultures were wounded in the presence of N-cadherin antibody (Figure 4a and 4b); furthermore, a high degree of localization of Cdc42-GFP to the cell anterior was suppressed (Figure 4d and 4e). In contrast, modest inactivation of RhoA at 1 hour postwounding was unaffected by the antibody (Figure 4c).
Cdc42 regulation of cell polarity of astrocytes and endothelium is effected via inhibition of GSK-3β.14,29 Accordingly, GSK-3β underwent rapid inhibitory phosphorylation after wounding of smooth muscle cultures that slowly decreased at later times (Figure 5a and 5b). In shear-stressed endothelium29 and in migrating astrocytes,15 GSK-3β inhibition must be spatially regulated to control cell polarity because wholesale inhibition of the kinase prevents MTOC reorientation. Accordingly, we found that 2 GSK-3β inhibitors, LiCl and SB415286, prevented the establishment of cell polarity in wound edge smooth muscle cells (Figure 5c). These results implicate N-cadherin–mediated inhibition of Cdc42 activity at the posterior–lateral borders of wound edge cells that organizes the microtubule network according to a Cdc42/GSK-3β signaling gradient.
Heterotypic Cell–Cell Contact Elicits N-Cadherin–Dependent Suppression of Smooth Muscle Cell Polarity
Arterial smooth muscle cells exhibit rapid migration to the intima and proliferation after loss of endothelium in vivo that is suppressed on reendothelialization. To test whether N-cadherin–specific adhesion inhibits directional migration of porcine arterial smooth muscle in already polarized cells, we exploited the A10 smooth muscle cell line, which does not express N-cadherin (Figure 6a) in heterotypic cell contact experiments. A10 cells stably expressing α-tubulin–GFP, to differentiate the cells from porcine arterial smooth muscle, were grown outside of cloning rings that enclosed porcine smooth muscle cell colonies, and then the barrier was removed allowing the 2 cell populations to meet. We then repeated the experiment using A10 cells that stably expressed N-cadherin–GFP. Interestingly, 6 hours after contact with N-cadherin–expressing A10 cells, a robust loss in the percentage of highly polarized cells was observed, whereas only a modest decrease occurred when primary smooth muscle cell contacted A10 cells that do not express N-cadherin (Figure 6b through 6f).
In vivo, smooth muscle cells in the innermost layer of the media contact endothelium, which expresses N-cadherin; therefore, we reasoned that N-cadherin signaling from endothelium may suppress polarity of these smooth muscle cells under resting conditions. To test this possibility in vitro, we grew endothelial cells or CHO cells (which do not express N-cadherin) outside of cloning rings that enclosed smooth muscle cell colonies and repeated the heterotypic contact experiments with these cell types. N-Cadherin antibody was used to distinguish CHO cells from smooth muscle cells, whereas a platelet endothelial cell adhesion molecule-1 antibody discriminated endothelial cells from smooth muscle cells. Forty percent of control cells at wound edges (no contact after same outward migration time, ≈36 hours) were highly polarized (quantitatively matched data from 6-hour wounds), whereas heterotypic contact with endothelium abolished polarity of smooth muscle cells (Figure 6b). When endothelium was replaced with CHO cells, which do not express N-cadherin, only a modest suppression of smooth muscle cell polarity occurred (Figure 6b). These results reveal that engagement of N-cadherin is a potent signal that abrogates cell polarity on wound closure.
N-Cadherin Does Not Regulate the Polarity of Migrating Endothelium
Wound-healing experiments were performed using postconfluent porcine aortic endothelial cells. As previously reported, these cells displayed a high degree of cell polarity at wound edges,12 but this polarity was not affected by treatment with N-cadherin antibody. Accordingly, at 6 hours after wounding, 81±10% of cells were highly polarized in the absence of N-cadherin–blocking antibody, and 69±3% were highly polarized when the antibody was present (P>0.05) (supplemental Figure III).
Angioplasty and stent implantation are attractive therapies for arterial occlusive disease because they are much less invasive than bypass procedures. The development of drug-eluting stents has been particularly encouraging because these devices suppress restenosis caused by intimal accumulation of vascular smooth muscle; however, recent concern has arisen over increased risk of late thrombosis with these devices.9 The likely source of late thrombosis is use of drugs that globally target cell proliferation and therefore suppress endothelial repair, as well as intimal smooth muscle accumulation. The present study focused on the early migration of smooth muscle cells to the luminal surface of the vessel as a possible target for restenosis therapy, because this process may be amenable to smooth muscle–specific therapies. We explored polarized signaling that innermost smooth muscle cells receive from underlying medial cells after loss of endothelium as potential stimuli for migration to the intima.
We examined the role of N-cadherin because it exhibits a polarized distribution to the posterior–lateral aspects of cells migrating at the leading edge of a cohort of cells and because its ligation can suppress activation of a key regulator of directional migration, Cdc42, in other cells.18 Such suppression has the potential to introduce the posterior-to-anterior gradient of Cdc42 activity in wound edge cells that drives polarity. We found that migrating smooth muscle cells also display polarity and that N-cadherin signaling can drive this critical cell behavior. The capacity of an N-cadherin antibody, an N-cadherin–targeted peptide, and calcium switch experiments to disrupt cadherin-dependent cell–cell contact leaves no doubt that the process requires N-cadherin–mediated signaling in smooth muscle cells.
The mechanisms that drive this polarity resemble those occurring in astrocytes and fibroblasts, including a dependence on intact microtubules (nocodazole experiments) and their dynamic instability (taxol experiments) and a dependence on the regulation of GSK-3β. The latter probably reflects tight spatial control of inhibition of the kinase that is initiated after injury because we and others have found that its wholesale inhibition with exogenous agents blocked polarity.15,29
In other cell types, integrin signaling at the leading edge of the cell, presumably as new cell substrate contacts are made, initiates polarity,17 but this anterior signaling is inadequate in smooth muscle cells because disruption of N-cadherin ligation fully ablated the process. It remains possible that signaling between N-cadherin and integrins could cooperate in driving cell polarity. For example, the nonreceptor tyrosine kinase Fer can shuttle between the cytoplasmic domains of N-cadherin and β1 integrins to modulate adhesion mediated by both complexes.30 The effects of this communication on cell polarity merit further investigation.
Endothelial cells engage in heterotypic cell contacts with smooth muscle cells through fenestrae of the internal elastic lamellae (our unpublished observations and see Spagnoli et al31); therefore, an attractive hypothesis was that endothelium stabilizes healthy arterial structure by presenting N-cadherin to innermost smooth muscle cells. Resulting signals would balance those from deeper smooth muscle and thereby suppress inward migration. Loss of endothelium, eg, during angioplasty, could then liberate a stimulus for inward migration. In fact, Paik et al have shown that N-cadherin–mediated interactions between endothelial cells and smooth muscle cells promotes vascular maturation and stability during development.19 In this study, we found that polarity of migrating smooth muscle cells is abolished specifically on contact with N-cadherin–expressing cells, including other smooth muscle cells and endothelium. These findings, combined with the effects of blocking N-cadherin adhesion, indicate that N-cadherin signaling is necessary for establishment of cell polarity following injury and a strong signal for loss of polarity on wound closure.
Although endothelial cells express appreciable levels of N-cadherin, there has been a consensus that homotypic endothelial adherens junctions are populated exclusively by VE-cadherin, which can displace N-cadherin from junctional complexes.32 Because of this displacement, polarized N-cadherin distribution in wound edge endothelial cells would not be anticipated. More recent in vivo work has challenged the universality of this concept33; nonetheless, we found that cell polarity of migrating endothelium was independent of N-cadherin ligation. The finding is important because it emphasizes the potential for disruption of N-cadherin ligation to selectively inhibit smooth muscle migration without affecting endothelial repair in a therapeutic setting.
All of the experiments in this study were performed using cultured cells grown on a 2D substrate, whereas arterial medial smooth muscle cells are normally embedded in a 3D extracellular matrix. Nonetheless, our findings are relevant to in vivo smooth muscle cell migration. This is because endothelial cell loss induces smooth muscle cells to crawl through fenestrae in the internal elastic lamellae, and they subsequently migrate laterally to cover the internal elastic lamellae. Consequently, the cells that lead this outgrowing cohort exhibit N-cadherin–mediated adhesion at the cell posterior, whereas the anterior is free of adherens junctions, because they migrate over a 2D surface.
Finally, we have emphasized the importance of directional migration of smooth muscle cells during restenosis, but the same concepts may apply to bypass graft failure and possibly to the migration of smooth muscle cells to atherosclerotic lesions. Furthermore, targeted migration of these cells or their precursors, sometimes over large distances, is central to developmental morphogenesis of the artery wall. Examples include the early investment of the proximal aorta, pulmonary trunk, and carotid arteries by cells originating in the cardiac neural crest, population of the distal aorta by cells derived from somites, and invasion of the heart by coronary pre cursor cells from the proepicardial organ (for review, see Majesky34). N-Cadherin is of proven importance in vessel wall assembly, and it will be of interest to explore its capacity to direct migration during vascular morphogenesis.
In summary, we have shown that smooth muscle cells display cell polarity during directional migration and that this polarity requires signals derived from N-cadherin–mediated adhesion and its regulation of a classical polarity pathway that involves Cdc42 and GSK-3β. The process is ablated by endothelial–smooth muscle cell contact and depends on N-cadherin expression in both cell types, because suppression of polarity is more efficient in the presence of cadherins. Failure of N-cadherin to suppress polarity of endothelial cells indicates that this molecule is a candidate target as a smooth muscle–specific mediator of intimal hyperplasia.
We are grateful to Dr Gregory Downey for providing the Cdc42-GFP construct and Dr Cecile Gauthier-Rouviere for providing the N-cadherin–GFP construct. Dr Mansoor Husain kindly provided the MOVAS cell line.
Sources of Funding
This work was supported by Heart and Stroke Foundation of Ontario grant NA 5332. M.P.B. is a Career Investigator of the Heart and Stroke Foundation. P.J.B.S. is a recipient of a Canada Graduate Scholarship Doctoral Award from the Canadian Institutes for Health Research.
Original received March 7, 2008; revision received June 25, 2008; accepted June 26, 2008.
Pasterkamp G, De Kleijn DPV, Borst C. Arterial remodeling in atherosclerosis, restenosis and after alteration of blood flow: potential mechanisms and clinical implications. Cardiovas Res. 2000; 45: 843–852.
Hao H, Gabbiani G, Bochaton PML. Arterial smooth muscle cell heterogeneity: implications for atherosclerosis and restenosis development. Arterioscler Thromb Vasc Biol. 2004; 23: 1510–1520.
Fingerle J, Au YPT, Clowes AW, Reidy MA. Intimal lesion formation in rat carotid arteries after endothelial denudation in absence of medial injury. Arteriosclerosis. 1990; 10: 1082–1087.
Jamal A, Bendeck MP, Langille BL. Structural changes and recovery of function after arterial injury. Arterioscler Thromb. 1992; 12: 307–317.
Jackson CL, Raines EW, Ross R, Reidy MA. Role of endogenous platelet-derived growth factor in arterial smooth muscle cell migration after balloon catheter injury. Arterioscler Thromb. 1993; 13: 1218–1226.
Serruys PW, Heyndrickx GR, Patel J, Cummins PA, Kleijne JA, Clowes AW. Effect of an anti-PDGF-β-receptor-blocking antibody on restenosis in patients undergoing elective stent placement. Acute Cardiac Care. 2003; 5: 214–222.
Daemen J, Serruys PW. Drug-eluting stent update 2007. Part II: Unsettled issues. Circulation. 2007; 116: 961–968.
Finn AV, Nakazawa G, Joner M, Kolodgie FD, Mont EK, Gold HK, Virmani R. Vascular responses to drug eluting stents. Importance of delayed healing. Arterioscler Thromb Vasc Biol. 2007; 27: 1500–1510.
Lin SX, Gundersen GG, Maxfield FR. Export from pericentriolar endocytic recycling compartment to cell surface depends on stable, detyrosinated (Glu) microtubules and kinesin. Mol Biol Cell. 2002; 13: 96–109.
Gotlieb AI, McBurnie May L, Subrahmanyan L, Kalnins VI. Distribution of microtubule organizing centers in migrating sheets of endothelial cells. J Cell Biol. 1981; 91: 589–594.
Etienne-Manneville S. Cdc42-the centre of polarity. J Cell Sci. 2004; 117: 1291–1300.
Schlessinger K, McManus EJ, Hall A. Cdc42 and noncanonical Wnt signal transduction pathways cooperate to promote cell polarity. J Cell Biol. 2007; 178: 355–361.
Charrasse S, Meriane M, Comunale F, Blangy A, Gauthier-Rouviere C. N-cadherin-dependent cell-cell contact regulates Rho GTPases and β-catenin localization in mouse C2C12 myoblasts. J Cell Biol. 2002; 158: 953–965.
Paik J-H, Skoura A, Chae S-S, Cowan AE, Han DK, Proia RL, Hla T. Sphingosine 1-phosphate receptor regulation of N-cadherin mediates vascular stabilization. Genes Dev. 2004; 18: 2392–2403.
Afroze T, Yang LL, Wang CS, Gros R, Kalair W, Hoque AN, Mungrue IN, Zhu ZP, Husain M. Calcineurin-independent regulation of plasma membrane Ca2+ ATPase-4 in the vascular smooth muscle cell cycle. Am J Physiol. 2003; 285: C88–C95.
Williams E, Williams G, Gour BJ, Blaschuk OW, Doherty P. A novel family of cyclic peptide antagonists suggests that N-cadherin specificity is determined by amino acids that flank the HAV motif. J Biol Chem. 2000; 275: 4007–4012.
Bendeck MP, Zempo M, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metaloproteinase expression after arterial injury in the rat. Circ Res. 1994; 75: 539–545.
Li G, Satyamoorthy K, Herlyn M. N-Cadherin-mediated intercellular interactions promote survival and migration of melanoma cells. Cancer Res. 2001; 61: 3819–3825.
El Sayegh TY, Arora PD, Laschinger CA, Lee W, Morrison C, Overall CM, Kapus A, McCulloch CA. Cortactin associates with N-cadherin adhesions and mediates intercellular adhesion strengthening in fibroblasts. J Cell Sci. 2004; 117: 5117–5131.
Jones M, Sabatini PJB, Lee FSH, Bendeck MP, Langille BL. N-Cadherin upregulation and function in response of smooth muscle cells to arterial injury. Arterioscler Thromb Vasc Biol. 2002; 22: 1972–1977.
McCue S, Dajnowiec D, Jackson MR, Langille BL. Shear stress regulates forward and reverse planar cell polarity of vascular endothelium in vivo and in vitro. Circ Res. 2006; 98: 939–946.
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.
Navarro P, Ruco L, Dejana E. Differential localization of VE- and N-cadherins in human endothelial cells: VE-cadherin competes with N-cadherin for junctional localization. J Cell Biol. 1998; 140: 1475–1484.
Luo Y, Radice GL. N-cadherin acts upstream of VE-cadherin in controlling vascular morphogenesis. J Cell Biol. 2005; 169: 29–34.
Majesky MW. Developmental basis of vascular smooth muscle diversity. Arterioscler Thromb Vasc Biol. 2007; 1; 27: 1248–1258.