Estrogen-Stimulated Endothelial Repair Requires Osteopontin
Objective— Estradiol (E2) is known to accelerate reendothelialization and thus prevent intimal thickening and in-stent restenosis after angioplasty. Transplantation experiments with ERα−/− mice have previously shown that E2 acts through local and bone marrow cell compartments to enhance endothelial healing. However, the downstream mechanisms induced by E2 to mediate endothelial repair are still poorly understood.
Methods and Results— We show here that after endovascular carotid artery injury, E2-enhanced endothelial repair is lost in osteopontin-deficient mice (OPN−/−). Transplantation of OPN−/− bone marrow into wild-type lethally irradiated mice, and vice versa, suggested that osteopontin plays a crucial role in both the local and the bone marrow actions of E2. In the vascular compartment, using transgenic mice expressing doxycyclin regulatable-osteopontin, we show that endothelial cell specific osteopontin overexpression mimics E2-enhanced endothelial cell migration and proliferation in the regenerating endothelium. In the bone marrow cell compartment, we demonstrate that E2 enhances bone marrow–derived mononuclear cell adhesion to regenerating endothelium in vivo, and that this effect is dependent on osteopontin.
Conclusions— We demonstrate here that E2 acceleration of the endothelial repair requires osteopontin, both for bone marrow–derived cell recruitment and for endothelial cell migration and proliferation.
Maintaining the integrity and function of the endothelial barrier is critical to vessel physiology. Indeed, endothelial damages lead to activation of thrombosis processes, namely after percutaneous transluminal coronary angioplasty often associated with stent fixing. It is now accepted that improving the reendothelialization process limits stent-induced thrombi formation.
See accompanying article on page 2099
Endothelial repair clearly involves migration and proliferation of endothelial cells (EC), which are in interaction with extracellular matrix proteins,1,2 as well as recruitment of bone marrow (BM)–derived cells including endothelial progenitor cells (EPC).3 At the molecular level, growth factors such as VEGF,4 FGF2,5 and compounds such as statins,3 prostacyclins,6 and estrogens7 accelerate endothelial repair. As previously demonstrated by us and others, 17β-estradiol (E2) accelerates endothelial healing via the estrogen receptor ER-α but not ER-β8 and through the involvement of endothelial NO synthase (eNOS)9,10 and basic FGF.11 This effect of E2 is mediated by BM cells, because it is lost in wild-type mice grafted with ERα-deficient BM cells.11 Moreover, E2 allows mobilization of EPCs12 and increases their number in the regenerating area of wounded arteries.9 On the other hand, E2 favors the proliferation and migration of ECs13 and prevents them from undergoing apoptosis.14 Despite these various effects, the molecular and cellular mechanisms that might account for the stimulating effects of E2 on endothelial repair remain unclear.7,15
Since the 1990s, estrogens have been known to regulate the synthesis of osteopontin (OPN).16 A first in vivo report demonstrated that in absence of OPN, mice are resistant to ovariectomy-induced bone resorption.17 Inhibition of OPN expression by estrogens was also found in vascular smooth muscle cells (SMCs) in vitro and in vivo.18 Conversely, estrogens are known to induce OPN expression in osteoblasts,19 in the kidney,20 and in the uterus luminal epithelium.21 Estrogen responsive elements were actually identified in the OPN promoter.22,23 Estrogen regulation of OPN promoter in ECs or EPCs is not known. OPN functions both as a cell attachment protein and as a chemokine, delivering signals to the cells via a number of receptors including several integrins and CD44.24 Moreover an intracellular form of OPN is also involved in cell adhesion and migration.25 As a cell attachment protein, OPN allows adhesion of numerous cell types including ECs.26 As a signaling molecule, OPN is able to modify gene expression and promote migration of various cell types including ECs, osteoclasts, monocytes,26 and SMCs.27 Moreover, OPN inhibits apoptosis in ECs28 and is involved in angiogenesis29 and postischemic neovascularization.30,31 In contrast to uninjured blood vessels where very little OPN is expressed, abundant levels of OPN were found in the remodeling media after balloon injury and in atherosclerosic plaques. OPN was also found at the regenerating endothelium wound edge26 and VEGF, which mediates acceleration of the reendothelialization, increased OPN expression in ECs,32 suggesting an involvement of OPN in endothelial repair. However, a recent report showed that in vitro, OPN overexpression dramatically impaired reendothelialization by inhibiting EC motility.33 Thus the role of OPN in vivo must be clarified.
Here, we have investigated the mechanisms involved in E2-stimulated endothelial repair and evaluated the implication of OPN in this process. We demonstrate that OPN is required for E2 to accelerate endothelial repair through cooperation between two cell compartments, ECs and BM-derived cells.
Materials and Methods
Mice were maintained in a conventional animal facility, on a 12-hour light/12-hour dark cycle. Food and water were available ad libitum. All procedures were carried out in compliance with the principles and guidelines established by the National Institute of Medical Research (INSERM) and approved by the Institutional Animal Care and Use Committee. Animals were anesthetized by an intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg).
OPN deficient mice (OPN−/−) are global knockout. Genotyping was performed as previously described.34 Tie2-rOPN mice (Tet off) were obtained by mating Tie2-tTA mice35,36 with TRE-rOPN mice carrying mouse recombinant OPN (rOPN) under the control of the doxycyclin/tetracycline regulatory element (TRE; please see supplemental materials, available online at http://atvb.ahajournals.org). This construct should allow coexpression of OPN and β-galactosidase with some differences (supplemental Figures I and II). Doxycyclin (7.8 mmol/L, Sigma) was added to the drinking water for 7 days before the experiment.
Animals were ovariectomized at 5 weeks of age and s.c. implanted in the back with either placebo or E2 releasing pellets (0.1 mg E2 delivering 80 μg/kg/d, Innovative Research of America). Carotid injury was generally performed 2 to 3 weeks after pellet implantation.
Bone Marrow Transplantation
Two weeks after ovariectomy, recipient mice were lethally γ-irradiated (9 Gy). Twenty-four hours later, they received an intravenous injection of 4×106 donor BM cells. One month later, mice were implanted with either placebo or E2 pellet.
Mouse Carotid Injury and Quantification of Endothelial Regeneration
Endovascular injury of the right carotid was performed mechanically using an in-house made swab (please see supplemental materials). No intimal thickening was observed after this endovascular injury.
Endothelial repair was evaluated by staining the denuded areas with Evans blue dye (Merck; please see supplemental materials) Percentage between blue stained area (unrepaired area) and injured carotid artery area was calculated.
Proliferating EC was evaluated by immunodetection of incorporated 5-bromo-2-deoxyuridine (please see supplemental materials). Proliferation index was calculated by dividing the number of proliferating ECs in ARA or RA by the respective surfaces of these areas.
OPN protein was detected using a goat polyclonal anti-OPN antibody (Sigma). ECs were identified using anti-CD31 antibody (BD Pharmigen), and nuclei were stained by the DAPI. The control antibody is a goat non immune IgG (Sigma).
In Vivo BM MNC Adhesion Test
The BM was flushed out of the cleaned bone (femur and tibia) with PBS. Mononuclear cells (MNCs) were isolated on a Ficoll gradient (Ficoll-Paque Plus, GE Healthcare, 1100 rpm, 30 minutes) and resuspended in EGM2 Bulletkit medium (Clonetics).
Two hours after carotid injury, mice received a retro-orbital injection of 4×106 PKH26-GL (Sigma) labeled BM MNC suspended in 250 μL of 150 mmol/L NaCl (please see supplemental materials). The recipient mouse carotids were harvested 24 hours later. The fluorescent BM MNCs adhering to the wounded area of the carotid artery were counted using fluorescent microscopy.
Results are expressed as means±SEM. Data were analyzed using Kruskal-Wallis 1-way ANOVA with NCSS statistical software. When the test indicated that at least 2 medians are different (probability value <0.05), a 2-sample t test (Mann–Whitney U) was performed to compare medians of each group to each other. A probability value <0.05 indicated significant difference between 2 groups. It was indicated by *. Nonsignificant difference was indicated by NS.
Endothelial OPN Contributes to Endothelial Repair
Previous conflicting in vitro reports show either that OPN is chemotactic for ECs26 or in contrast that it inhibits in vitro reendothelialization in a model of cultured EC layer injury.33 To clarify this point, we have generated Tie2-rOPN transgenic mice designed to overexpress recombinant OPN in ECs in a doxycyclin-regulatable fashion (Tet-Off strategy). These mice allowed to evaluate the role of EC OPN expression during in vivo EC migration occurring during the endothelial repair after experimental endothelial injury of the mouse carotid artery.
In Tie2-rOPN mice, immunohistological analyses showed that OPN protein is expressed in all the ECs of the healthy carotid artery while it is not detected in healthy carotid artery of wild-type mice (supplemental Figure II). In these mice, the endothelial repair was accelerated compared to that in Tie2-rOPN mice in which the transgene expression was inhibited by doxycyclin (+Dox) (Figure 1). These data demonstrate that in vivo, OPN overexpression in ECs induced an acceleration of the endothelial repair and thus suggest that in vivo, OPN increases EC migration.
E2-Accelerated Reendothelialization Requires OPN
As previously reported,8 E2 treatment of ovariectomized wild-type mice induced a significant increase of the repaired endothelial area measured 3 and 5 days after injury compared to placebo-treated mice (supplemental Figure IIIA; Figure 2A). Interestingly, the E2 effect was lost in OPN-deficient (OPN−/−) mice. Because the endothelial repair was identical in OPN+/+ and OPN−/− placebo-treated mice (Figure 2A), this experiment demonstrated that OPN is required for the E2-mediated acceleration of endothelial repair.
When OPN expression was rescued in EC of OPN−/− mice expressing rOPN (Tie2-rOPN x OPN−/− mice in absence of doxycyclin), endothelial healing was accelerated compared to that in control doxycyclin-watered mice (Figure 1). Endothelial OPN overexpression, alone, was thus sufficient to induce an acceleration of the endothelial repair. This led us to postulate that OPN is a crucial downstream mediator of E2 effects on endothelial healing.
To characterize events involved in E2-mediated endothelial repair, proliferating ECs were identified and counted. The number of proliferating ECs in the injured carotid 5 days after injury was significantly increased in E2-treated compared to placebo-treated OPN+/+ mice (Figure 2B). Proliferating ECs were found not only in the repaired endothelium area (RA) but also in a retrograde area (ARA) located in the healthy endothelium upstream of the lesion site. Respective surfaces of these 2 areas were strongly increased in E2-treated OPN+/+ mice (Figure 2C; supplemental Figure IIIB and IIIC). In the RA, the E2-mediated increase in proliferating ECs was proportional to the increase of its surface. The proliferating EC index was thus not significantly different with placebo or E2 treatment (Figure 2D). In contrast, in the ARA, the proliferation index was significantly increased in E2-treated OPN+/+ mice (Figure 2D).
In OPN−/− mice, E2 was unable to increase the number of proliferating ECs, or to increase the ARA size, which were similar to those in OPN+/+ mice with placebo treatment (Figure 2B and 2C).
These results thus demonstrate that the E2-mediated processes leading to the acceleration of endothelial repair absolutely require OPN.
Origin of OPN Required for the Effect of E2 on Endothelial Healing
The OPN required for the E2-induced endothelial repair increase could come from either vascular or nonvascular cells. OPN was not detected in any layer of the healthy carotid artery of wild-type mice (supplemental Figure II). In contrast, 3 days after endothelium removal, OPN protein was detected in SMCs of the media in the vicinity of the carotid artery bifurcation and in inflammatory cells present in the adventitia, mainly at the level of areas that remained deendothelialized (Figure 3, right panel). Interestingly, OPN was also detected in ECs of the endothelial repaired area but not in ECs of the uninjured endothelial area far from the limit of injury (Figure 3).
Because we previously had shown BM-derived cells to be involved in the E2 effect on reendothelialization,11 we have studied OPN expression in cultured BM-derived cells and demonstrate here that these cells not only expressed the OPN protein (supplemental Figure IVA) but also that the level of OPN expression was dependent on E2 activity (supplemental Figure IVB).
We further tested the involvement of OPN from either BM-derived or vascular cells in E2 effect. We therefore generated hematopoietic chimeric mice by grafting OPN−/− BM to lethally irradiated OPN+/+ mice and vice versa. One month after grafting, endothelial repair was assessed 3 days after endothelial injury. In control mice (OPN+/+ BM≫ OPN+/+ mouse), E2 still induced an acceleration of endothelial repair, showing that neither the irradiation nor the BM transplantation altered the E2 effect (Figure 4). When OPN+/+ BM was grafted in OPN−/− mice, E2 was unable to accelerate the endothelial repair, suggesting that OPN from BM-derived cells is not sufficient to mediate the E2 effect. Similarly, when OPN−/− BM was grafted in OPN+/+ mice, E2 was also unable to induce an acceleration of the endothelial repair (Figure 4). These results suggest that OPN expression in both BM-derived cells and the remaining vascular cells is required to bring about the E2 effect on reendothelialization.
OPN Is Involved in the E2-Induced Increase of BM-Derived Cell Adhesion on the Regenerating Endothelial Area
Although OPN is required in E2-mediated reendothelialization, and E2 supplementation increases EPC number, results of supplemental Figure V indicated that the loss of the E2 effect on reendothelialization in OPN−/− mice was not the result of an alteration of the mobilization of this BM-derived cell population.
According to the adhesive property of OPN,37 we hypothesized that the OPN contribution to reendothelialization could be mediated through its capacity to enhance the adhesion of BM-derived cells to the injured carotid artery in the presence of E2. We therefore investigated the ability of PKH26-GL-labeled BM MNCs from OPN+/+ or OPN−/− mice to adhere to the injured carotid artery of OPN+/+ or OPN−/− ovariectomized mice treated with E2 or placebo. We showed first that OPN−/− BM MNCs adhered to a lesser extent than OPN+/+ BM MNCs to OPN+/+ mouse injured carotid artery (Figure 5). Moreover a decrease in OPN+/+ BM MNC adhesion to OPN−/− injured carotid artery was observed. Finally we tested the E2 effect and demonstrated that BM MNC adhesion was enhanced by E2 supplementation only when both MNCs and vascular cells expressed OPN (Figure 5). Altogether these data demonstrate that OPN is involved in BM-derived cell adhesion in vivo and is required for increased adherence induced by E2 treatment.
We report here for the first time that E2 requires the multifunctional protein OPN to mediate acceleration of endothelial repair after injury. Moreover, we show that this is a complex process requiring vascular as well as BM-derived events because OPN from both bone marrow and nonbone marrow origins is involved in this process.
Reendothelialization is mainly the result of migration and proliferation of ECs located at the edges of the lesion site in the rat aorta.1 In the mouse carotid, the reendothelialization process occurs mainly from the proximal edge of the injured endothelium, in the direction of the blood flow.2 Previous reports suggested that, in vitro, E2 was able to increase EC migration and proliferation.13 We demonstrate here that, in vivo, E2 is able to increase the EC proliferation index not only in the regenerating area (RA) but also in a healthy endothelial area located upstream of the lesion (ARA), thus suggesting that the increased rate of endothelial repair resulted from an enhanced EC migration toward the acellular area.2
In this article we demonstrate that E2-mediated enhancement of EC migration and proliferation is dependent on OPN. Indeed E2 was unable to induce these effects in OPN−/− mice. Thus, OPN could be considered as a downstream component of the E2 pathway in EC.
Since 1995, OPN was recognized as a chemotactic factor for ECs.26 A controversy regarding the role of OPN originated from the work of Leali et al,33 which showed that OPN overexpression inhibited in vitro reendothelialization. Our data suggest that OPN is not required for E2-independent in vivo reendothelialization, because the velocity of endothelial repair was identical in OPN+/+ and OPN−/− mice under placebo treatment but is required for E2-mediated increase of endothelial repair. The fact that OPN deficiency did not affect endothelial repair suggested that an E2-independent compensatory mechanism could occur in OPN−/− mice. When OPN was constitutively expressed in an EC-specific manner in the TET-inducible transgenic mouse model, reendothelialization was accelerated, suggesting that in vivo, OPN overexpression indeed stimulates endothelial repair. This result underlines the difference between in vivo and in vitro studies probably linked to the differences in EC phenotype, in the extracellular matrix layer and potentially at the level of OPN expression in ECs.
Bone marrow transplantation experiments demonstrating that OPN, whether from the bone marrow or the local compartments, is unable to assume alone the E2-mediated improvement of the endothelial repair, suggested an autocrine role for OPN. Because the intracellular form of OPN has been previously involved in osteoblast and fibroblast migration,25 we can also hypothesize that in our in vivo model, OPN would act intracellularly in ECs rather than through an extracellular action involving integrin engagement.33
Neointima formation after acute endovascular injury is inhibited by E2 supplementation and different mechanisms have been proposed including inhibition of SMC migration and proliferation, and stimulation of reendothelialization.12,38 Interestingly, it was suggested by an in vitro study that E2 inhibits OPN expression in SMCs, which is required for adventitial fibroblast migration involved in intimal thickening.18 In addition, our data show that E2 mediates the acceleration of endothelial repair through an OPN-dependent pathway suggesting dual opposite roles for OPN in SMCs and ECs, which both contribute to neointima inhibition.
Different lines of evidence show that E2 increased the number of EPCs in BM and blood but also in the regenerating endothelium.9,12 We show here that E2 not only increases BM mobilization but also enhances their homing to the injured area, and that OPN is required in this last effect. Mechanisms involved in the homing are complex, but adhesion molecules previously known to be involved in phases of rolling and firm adhesion of leukocytes appear as key regulators. Here we suggest that OPN is also a key factor of the BM cell homing. How OPN is involved remains unclear, but this effect is probably linked to the OPN ability to bind on one side to integrin and CD44 receptors and on the other side to extracellular matrix components, notably to collagen.
Altogether, these results suggest a possible cascade of events occurring during E2-mediated acceleration of endothelial repair (Figure 6). In a first step, E2 increases the pool of BM-derived cells according to a mechanism involving FGF211 but not OPN. Then, E2 induces the production of OPN by BM-derived cells, allowing them to more efficiently adhere to endothelial regenerating areas. The role of adherent BM-derived cells is still unknown, but they could contribute to an OPN-dependent E2-mediated stimulation of EC migration from the adjacent healthy zone. Finally, we can speculate that consequently to the EC migration from the healthy adjacent endothelium (ARA) toward the deendothelialized area, the cell density decreases in ARA, leading to an EC proliferation to refill an adequate cell density.
We thank Jérôme Guignard for technical assistance in the animal facility, Isabelle Belloc for her excellent technical assistance, and Philippe Alzieu for development of the endovascular injury model. TRE-rOPN transgenic mice were created and expanded by Pierre Costet and Franck Simonnet at the University of Bordeaux 2 transgenic animal facility.
Sources of Funding
This study was supported by grants from Conseil Régional d’Aquitaine (action inter-régionale Aquitaine Midi Pyrénées), Communauté de Travail des Pyrénées, ANR (A05109GS), Fondation de France (program santé), and the European Vascular Genomics Network (www.evgn.org), a Network of Excellence supported by the European Community’s sixth Framework Programme for Research Priority “Life sciences, genomics and biotechnology for health” (contract LSHM-CT-2003-503254). L.L.S.L. and F.R. were recipients of fellowships from the French Research Ministry and the Aquitaine Region, respectively.
Original received April 4, 2008; final version accepted August 25, 2008.
Filipe C, Lam Shang Leen L, Brouchet L, Billon A, Benouaich V, Fontaine V, Gourdy P, Lenfant F, Arnal JF, Gadeau AP, Laurell H. Estradiol accelerates endothelial healing through the retrograde commitment of uninjured endothelium. Am J Physiol. 2008; 294: H2822–H2830.
Werner N, Priller J, Laufs U, Endres M, Bohm M, Dirnagl U, Nickenig G. Bone marrow-derived progenitor cells modulate vascular reendothelialization and neointimal formation: effect of 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibition. Arteiroscler Thromb Vasc Biol. 2002; 22: 1567–1572.
Asahara T, Bauters C, Pastore C, Kearney M, Rossow S, Bunting S, Ferrara N, Symes JF, Isner JM. Local delivery of vascular endothelial growth factor accelerates reendothelialization and attenuates intimal hyperplasia in balloon-injured rat carotid artery. Circulation. 1995; 91: 2793–2801.
Numaguchi Y, Naruse K, Harada M, Osanai H, Mokuno S, Murase K, Matsui H, Toki Y, Ito T, Okumura K, Hayakawa T. Prostacyclin synthase gene transfer accelerates reendothelialization and inhibits neointimal formation in rat carotid arteries after balloon injury. Arterioscler Thromb Vasc Biol. 1999; 19: 727–733.
Brouchet L, Krust A, Dupont S, Chambon P, Bayard F, Arnal JF. Estradiol accelerates reendothelialization in mouse carotid artery through estrogen receptor-alpha but not estrogen receptor-beta. Circulation. 2001; 103: 423–428.
Iwakura A, Luedemann C, Shastry S, Hanley A, Kearney M, Aikawa R, Isner JM, Asahara T, Losordo DW. Estrogen-mediated, endothelial nitric oxide synthase-dependent mobilization of bone marrow-derived endothelial progenitor cells contributes to reendothelialization after arterial injury. Circulation. 2003; 108: 3115–3121.
Billon A, Lehoux S, Lam Shang Leen L, Laurell H, Filipe C, Benouaich V, Brouchet L, Dessy C, Gourdy P, Gadeau AP, Tedgui A, Balligand JL, Arnal JF. The estrogen effects on endothelial repair and mitogen-activated protein kinase activation are abolished in endothelial nitric-oxide (NO) synthase knockout mice, but not by NO synthase inhibition by N-nitro-L-arginine methyl ester. Am J Pathol. 2008; 172: 830–838.
Fontaine V, Filipe C, Werner N, Gourdy P, Billon A, Garmy-Susini B, Brouchet L, Bayard F, Prats H, Doetschman T, Nickenig G, Arnal JF. Essential role of bone marrow fibroblast growth factor-2 in the effect of estradiol on reendothelialization and endothelial progenitor cell mobilization. Am J Pathol. 2006; 169: 1855–1862.
Strehlow K, Werner N, Berweiler J, Link A, Dirnagl U, Priller J, Laufs K, Ghaeni L, Milosevic M, Bohm M, Nickenig G. Estrogen increases bone marrow-derived endothelial progenitor cell production and diminishes neointima formation. Circulation. 2003; 107: 3059–3065.
Geraldes P, Sirois MG, Tanguay JF. Specific contribution of estrogen receptors on mitogen-activated protein kinase pathways and vascular cell activation. Circ Res. 2003; 93: 399–405.
Spyridopoulos I, Sullivan AB, Kearney M, Isner JM, Losordo DW. Estrogen-receptor-mediated inhibition of human endothelial cell apoptosis. Estradiol as a survival factor. Circulation. 1997; 95: 1505–1514.
Yoshitake H, Rittling SR, Denhardt DT, Noda M. Osteopontin-deficient mice are resistant to ovariectomy-induced bone resorption. Proc Natl Acad Sci U S A. 1999; 96: 8156–8160.
Li G, Chen YF, Kelpke SS, Oparil S, Thompson JA. Estrogen attenuates integrin-beta(3)-dependent adventitial fibroblast migration after inhibition of osteopontin production in vascular smooth muscle cells. Circulation. 2000; 101: 2949–2955.
White FJ, Ross JW, Joyce MM, Geisert RD, Burghardt RC, Johnson GA. Steroid regulation of cell specific secreted phosphoprotein 1 (osteopontin) expression in the pregnant porcine uterus. Biol Reprod. 2005; 73: 1294–1301.
Vanacker JM, Pettersson K, Gustafsson JA, Laudet V. Transcriptional targets shared by estrogen receptor- related receptors (ERRs) and estrogen receptor (ER) alpha, but not by ERbeta. EMBO J. 1999; 18: 4270–4279.
Zirngibl RA, Chan JS, Aubin JE. Estrogen receptor-related receptor alpha (ERRalpha) regulates osteopontin expression through a non-canonical ERRalpha response element in a cell context-dependent manner. J Mol Endocrinol. 2008; 40: 61–73.
Liaw L, Lindner V, Schwartz SM, Chambers AF, Giachelli CM. Osteopontin and beta 3 integrin are coordinately expressed in regenerating endothelium in vivo and stimulate Arg-Gly-Asp-dependent endothelial migration in vitro. Circ Res. 1995; 77: 665–672.
Chaulet H, Desgranges C, Renault MA, Dupuch F, Ezan G, Peiretti F, Loirand G, Pacaud P, Gadeau AP. Extracellular nucleotides induce arterial smooth muscle cell migration via osteopontin. Circ Res. 2001; 89: 772–778.
Duvall CL, Weiss D, Robinson ST, Alameddine FM, Guldberg RE, Taylor WR. The role of osteopontin in recovery from hind limb ischemia. Arterioscler Thromb Vasc Biol. 2008; 28: 290–295.
Leali D, Moroni E, Bussolino F, Presta M. Osteopontin overexpression inhibits in vitro re-endothelialization via integrin engagement. J Biol Chem. 2007; 282: 19676–19684.
Dufourcq P, Leroux L, Ezan J, Descamps B, Daniel Lamazière JM, Costet P, Basoni C, Moreau C, Deutsch U, Couffinhal T, Duplàa C. Regulation of endothelial cell cytoskeletal reorganization by a sFRP-1 and Frizzled 4 and 7 dependent pathway; role in neo vessel formation. Am J Pathol. In press.
Liaw L, Allmeida M, Hart CE, Schwartz SM, Giachelli CM. Osteopontin promotes vascular cell adhesion and spreading and is chemotactic for smooth muscle cells in vitro. Circ Res. 1994; 74: 214–224.