Caveolin Plays a Central Role in Endothelial Progenitor Cell Mobilization and Homing in SDF-1–Driven Postischemic Vasculogenesis
http://www.100md.com
Elhem Sbaa, Julie DeWever, Philippe Mart
参见附件。
the Unit of Pharmacology and Therapeutics, University of Louvain Medical School, Brussels, Belgium.
Abstract
When neovascularization is triggered in ischemic tissues, angiogenesis but also (postnatal) vasculogenesis is induced, the latter requiring the mobilization of endothelial progenitor cells (EPC) from the bone marrow. Caveolin, the structural protein of caveolae, was recently reported to directly influence the angiogenic process through the regulation of the vascular endothelial growth factor (VEGF)/nitric oxide pathway. In this study, using caveolin-1 null mice (Cav–/–), we examined whether caveolin was also involved in the EPC recruitment in a model of ischemic hindlimb. Intravenous infusion of Sca-1+ Lin– progenitor cells, but not bone marrow transplantation, rescued the defective neovascularization in Cav–/– mice, suggesting a defect in progenitor mobilization. The adhesion of Cav–/– EPC to bone marrow stromal cells indeed appeared to be resistant to the otherwise mobilizing SDF-1 (Stromal cell–Derived Factor-1) exposure because of a defect in the internalization of the SDF-1 cognate receptor CXCR4. Symmetrically, the attachment of Cav–/– EPC to SDF-1–presenting endothelial cells was significantly increased. Finally, EPC transduction with caveolin small interfering RNA reproduced this advantage in vitro and, importantly, led to a more extensive rescue of the ischemic hindlimb after intravenous infusion (versus sham-transfected EPC). These results underline the critical role of caveolin in ensuring the caveolae-mediated endocytosis of CXCR4, regulating both the SDF-1–mediated mobilization and peripheral homing of progenitor cells in response to ischemia. In particular, a transient reduction in caveolin expression was shown to therapeutically increase the engraftment of progenitor cells.
Key Words: caveolin caveolae postnatal vasculogenesis endothelial progenitor cells ischemia
Introduction
Vasculogenesis and angiogenesis are the 2 major processes contributing to neovascularization in adults.1,2 Postnatal vasculogenesis in adults requires the recruitment of progenitor cells derived from the bone marrow,3 whereas angiogenesis involves the proliferation and migration of endothelial cells from the preexisting vessels.4 The proportion of either process in promoting the development of neovessels in pathological situations, such as ischemia and tumor growth, may vary and both are considered to be complementary.3,5 Moreover, the 2 processes share biological mediators including chemocytokines and proteinases.6
Among the key players in postnatal vasculogenesis, SDF-1 (Stromal cell–Derived Factor-1) and vascular endothelial growth factor (VEGF) have undoubtedly been the focus of the most attention. Through specific interactions with the CXCR4 and VEGFR-2 receptors, SDF-1 and VEGF, respectively, contribute to increase the recruitment of progenitor cells from the bone marrow to the peripheral blood, thereby leading to the formation of new vessels at the ischemic sites.2 Recently, we identified a critical role of caveolin-1 (a structural protein of caveolae) in the regulation of the VEGF-signaling pathway in a model of adaptive angiogenesis.7 We found that the compartmentation of the VEGFR-2 receptor in caveolae was key to ensure VEGF-mediated neovascularization. Accordingly, genetic deletion of caveolin-1 (and the ensuing absence of caveolae) was found to alter the endothelial nitric oxide synthase (eNOS) and ERK-dependent pathways downstream of VEGF stimulation, thereby leading to a defect in angiogenesis. A key role of caveolin-1 in angiogenesis was also identified in mouse tumor models and in the implanted Matrigel plug assay,8–10 mostly through the modulation of nitric oxide bioavailability. Notably, in none of these studies focusing on the role of caveolin in neovascularization was the potential impact of caveolin/caveolae in postnatal vasculogenesis (versus angiogenesis) addressed.
A series of indirect observations, however, argue for a role of caveolin in the bone marrow cell–dependent neovessel formation: (1) the activity of matrix metalloprotease-9 (MMP-9), a major protease involved in stem cell mobilization is modulated by nitric oxide,11 the production of which is directly dependent on caveolin abundance12; (2) some proteins involved in the interaction between progenitor cells and bone marrow stroma, including adhesion molecules13–15 and chemokine receptor,16 have already been associated to caveolae in other tissues; (3) the quasilinear relationship between caveolin expression and caveolae density,17 and the capacity of caveolae to pinch off and to sequester their content18 may account for the dynamic regulation of the expression of 1 or several of the abovementioned proteins at the cell surface.
Such evidence suggesting the potential role of caveolin/caveolae in progenitor cell biology and mobilization prompted us to explore the vasculogenic process in caveolin-deficient mice. More generally, another purpose of this study was to determine whether promotion of postnatal vasculogenesis could help rescue ischemic tissues in caveolin-null mice. Alteration in caveolin abundance has, indeed, been reported in many cardiovascular diseases that themselves are associated with a defect in angiogenesis.19,20
Materials and Methods
An expanded Materials and Methods section is provided in the online data supplement at http://circres.ahajournals.org.
Animals
Caveolin-deficient mice (Cav–/–)21 and their control littermates (Cav+/+) were generated through heterozygous matings and housed in our local facility. Anesthetized mice underwent a double femoral artery and vein ligation to allow resection of 2-mm vessel segments under a stereoscopic microscope while innervation was carefully preserved.
Bone Marrow and Endothelial Progenitor Cells
Bone marrow mononuclear cells were isolated by separation onto a Ficoll Histopaque gradient (Sigma), and magnetic cell sorting (Miltenyi Biotech) was used for positive and/or negative selection to collect Sca-1+/lineage marker-negative (Lin–) cells or CD45+ cells (BD Bioscience Pharmingen). For endothelial progenitor cell (EPC) production, bone marrow mononuclear cells were directly cultured on fibronectin-coated flasks using endothelial cell growth medium, as previously described.22 The identity of EPC was confirmed by morphological and immunocytochemical analyses. EPC were used at day 7 after initial seeding except when indicated otherwise.
Results
Infusion of Sca-1+ Lin– Bone Marrow Cells Rescues Ischemic Hindlimb Perfusion in Cav–/– Mice
We have previously reported that after resection of 2 mm-femoral artery/vein segments, Cav–/– mice never restored a normal hindlimb perfusion, compared with wild-type mice that completely recovered7 (see also below). To evaluate whether progenitor cells could help rescuing the impaired neovascularization in Cav–/– mice, we examined the effects of the infusion of Sca-1+ Lin– BMC on the restoration of the perfusion in the operated hindlimb. Serial examinations of blood flow by laser Doppler imaging were performed on days 3, 7, 10, and 14 (day 0 being fixed as the day of ligature). Figure 1A shows that infusion of 106 BMC issued either from wild-type (Cav+/+) or Cav–/– mice did rescue the limb perfusion, whereas a permanent drop in blood flow was observed in the operated leg of untreated Cav–/– mice. Of note, in Cav+/+ mice, infusion of Cav+/+ BMC barely influenced the extent (and rate) of the otherwise naturally occurring reperfusion of the operated hindlimb (see Figure 1B).
In some experiments, we loaded BMC with a long-lasting tracker (carboxymethyl-1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine [CM-DiI]) to follow their in vivo fate after infusion. Accordingly, histological analyses confirmed the presence of CM-DiI–labeled cells in the adductor muscle of Cav–/– mice, regardless of the Cav–/– or Cav+/+ mouse origin of infused cells (Figure 1C). CM-DiI–labeled cells could be found as early as 24 hours after induction of ischemia and were still detectable at day 14 (at much lower level). It should, however, be noted that posttransplantation dilution of the tracker (with cell division) may lead to underestimation of progenitor cell recruitment at longer times. The CM-DiI overlapping with CD31-positive capillary endothelial cells amounted to 29±12% and 37±13% (of total CM-DiI staining) in experiments with Cav–/– and Cav+/+ BMC, respectively (Figure 1C).
Ischemia-Induced Mobilization of Cav–/– Progenitor Cells Is Defective
Because infusion of Cav–/– BMC in Cav–/– mice did restore normal blood flow, a defect in mobilization could be suspected in these mice. We, therefore, set up a series of experiments where bone marrow CD45+ cells (eg, depleted of stromal cells) from Cav+/+ and Cav–/– mice were transplanted into lethally irradiated Cav–/– mice; Cav+/+ mice were similarly transplanted to allow full comparison between the different combinations of donor and recipient mice. A 3-week delay was respected after transplantation to ensure the complete reconstitution of bone marrow; preliminary studies had shown that circulating leukocyte number and bone marrow integrity were back to normal 14 days after CD45+ cell infusion. Figure 2A, left panels, shows that Cav–/– mice recovered from the ischemic insult when CD45+ cells from wild-type mice, but not from Cav–/– mice, were used to reconstitute the bone marrow. In Cav+/+ background animals (Figure 2A, right panels), the reconstitution of bone marrow also led to distinct patterns of response according to the origin of CD45+ cells, although to a lesser extent than in Cav–/– animals. Accordingly, when Cav–/– CD45+ cells were used, the restoration of blood flow in the proximal operated limb was delayed when compared with wild-type CD45+-reconstituted mice (see Figure 2A, right panels). In fact, whereas in the latter mice, no distal limb atrophy was detectable, the loss of toes or the entire foot of the operated leg was consistently observed in Cav–/– CD45+-reconstituted wild-type mice. These morphological alterations were less prominent, however, than in Cav–/– CD45+-reconstituted Cav–/– mice; a complete leg atrophy occurred in a majority of these mice (5 of 6).
Based on the above results, besides the transplanted mouse background, the origin of bone marrow seemed also critical for postischemic revascularization (eg, no or delayed recovery with CD45+ cells originated from Cav–/– mice). We therefore repeated the above experiments by using CM-DiI–labeled CD45+ cells to reconstitute the bone marrow of irradiated mice (see Figure I in the online data supplement) and to further track the progenitor cell homing in the ischemic hindlimb. In Cav–/– mice (Figure 2B, left panels), progenitor cells derived from the Cav–/– CD45+ cell transplantation were found heterogeneously distributed across the ischemic muscle and, in total, much less abundant than those arising from Cav+/+ cell transplantation (see Figure 2C for quantitative analysis). In Cav+/+ mice, although similar extents of CM-DiI–labeled progenitor cells issued either from Cav+/+ or Cav–/– mice could be found in some regions of the adductor muscle (see Figure 2B, right panels), the distribution was less homogeneous in the latter condition. Accordingly, quantitative analyses of the all tissue sections revealed a lesser staining when progenitor cells were issued from Cav–/– CD45+ cell transplantation (see bars 3 and 4 in Figure 2C).
The microvascular density (see CD31 immunolabeling in Figure 2B, lower panels) was significantly higher in both mouse strains when the CD45+ donor mouse was from a wild-type background (see Figure 2D for quantitative analysis); colocalization of CM-DiI and CD31 was however very poor suggesting paracrine influence of progenitor cells instead of direct participation to the structure of new capillaries. Moreover, slight but statistically significant differences distinguish the animals from Cav+/+ and Cav–/– backgrounds: (1) the extent of CM-DiI–labeled Cav+/+ progenitor cells found in the ischemic muscle was larger in Cav–/– recipient (compare bars 2 and 4 in Figure 2C); (2) the extent of CD31–labeled cells was larger in Cav+/+ background animals (compare bars 1 and 3 in Figure 2D).
Cav–/– EPC Are Resistant to SDF-1–Induced Detachment From Bone Marrow Stromal Cells
To further dissect the mechanism accounting for the defect in mobilization of Cav–/– BMC, we examined whether the MMP-9 and SDF-1 pathways were differently regulated in wild-type and caveolin-deficient mice. Zymography analyses of crude bone marrow extracts did not reveal significant differences in gelatinase activity (not shown), suggesting the absence of alterations in MMP-2 and -9 activities in Cav–/– mice. Furthermore, because we were anticipating alterations in the EPC phenotype more than in the bone marrow stroma (where most of MMP activity originates), we focused our attention on SDF-1/CXCL12 and its receptor CXCR4 located on EPC. Accordingly, we isolated stromal cells from Cav–/– and Cav+/+ by negative selection (CD45–) from total bone marrow and cultured them to confluence in the presence of serum. Stromal cell monolayers were then exposed to hypoxia (1% O2 for 4 hours) to promote endogenous SDF-1 expression as previously documented.23 Adhesion experiments were performed with EPC originating from both mouse genotypes, prechallenged or not with SDF-1. Figure 3A shows that when EPC were issued from Cav+/+ mice, the SDF-1 treatment led to a decreased ability to adhere to stromal cells independently of their genotype (–85±7% and –72±10% with stromal cells from Cav+/+ and Cav–/–, respectively, versus untreated EPC; P<0.01, n=3). Conversely, Cav–/– EPC were resistant to SDF-1 stimulation and did not reveal significant reduction in adhesion when compared with untreated EPC (see also Figure 3B for quantitative analysis). In the same experimental set up, however, CXCR4 antibody blockade could abolish the Cav+/+ and Cav–/– EPC adhesion on Cav–/– stromal cells to the same extent, further confirming the implication of the SDF-1/CXCR4 axis in the caveolin-mediated mobilization of EPC (Figure 3B). Fluorescence-activated cell sorting (FACS) analysis was then performed with EPC exposed or not to SDF-1 to examine the level of cell surface expression of the CXCR4 receptor, the SDF-1 cognate receptor. Although the total amounts of CXCR4 receptors were not significantly different in untreated Cav+/+ and Cav–/– EPC, the SDF-1 treatment did induce a shift to the left of the fluorescent peak in Cav+/+ EPC, indicating the disappearance of part of CXCR4 receptors from the cell surface (Figure 3C). Such shift in fluorescence was not observed with SDF-1–treated Cav–/– EPC, thereby illustrating the resistance to internalization of CXCR4 when caveolin was absent (Figure 3C). We also performed CXCR4 immunoblotting on detergent-insoluble membranes enriched in rafts and caveolae, from both Cav+/+ and Cav–/– EPC (Figure 3D). We found that in EPC from both origins, SDF-1 similarly induced the translocation of a large proportion of CXCR4 receptors to the detergent-insoluble fractions (approximately +40% of total CXCR4 receptors; see the shift between lanes 1 and 2 in Figure 3D). The implication of clathrin-coated pits was further excluded because the use of the cationic amphiphilic drug chlorpromazine, an inhibitor of clathrin-dependent, receptor-mediated endocytosis, failed to affect the CXCR4 internalization (Figure 3D). More importantly, when caveolin immunoprecipitation was performed to remove caveolar membranes (eg, to discriminate from rafts), the CXCR4 receptor could not be found in the remaining supernatant of Cav+/+ EPC membranes, confirming the key role of caveolae in CXCR4 sequestration (see lane 4 in Figure 3D).
Ischemia Leads to an Increase in BMC Caveolin Expression and in Plasma SDF-1 Levels
Because the data presented above indicated a critical role of caveolin and SDF-1 in EPC mobilization, we examined how their expression evolved with time in situ. Accordingly, we used real time PCR to first compare the extent of caveolin-1 mRNA transcripts in Sca-1+ Lin– BMC collected at different time points after induction of hindlimb ischemia. Figure 4A shows that although caveolin-1 is almost absent from these hematopoietic cells in normal conditions, its expression rapidly increased in response to peripheral ischemia to reach a maximum after 7 days. The elevated level of caveolin expression in Sca-1+ Lin– BMC was maintained for 2 more weeks before rapidly dropping back to basal values. Finally, we examined whether cultured EPC also exhibited changes in caveolin abundance following in vitro exposure to serum and VEGF. Figure 4B shows that, indeed, the level of caveolin progressively increased to reach a maximum at day 14 of culture. Also, evaluation of circulating levels of SDF-1 revealed that in basal conditions, plasma levels of SDF-1 were 2-fold higher in Cav–/– than in Cav+/+ mice. Following ischemic insult, the SDF-1 plasma levels increased in Cav+/+ mice to reach the level of Cav–/– mice (Figure 4C). Interestingly, however, the amounts of circulating EPC (determined by numeration of colonies) appeared to be 41±19% higher in Cav+/+ mouse plasma (versus Cav+/+ mice; P<0.01, n=7).
Lack of Caveolin Increases the EPC Adhesion to Endothelial Cells
We next examined whether alterations in the caveolin-mediated CXCR4 internalization, which translated in a defect in mobilization from the bone marrow, also impacted the peripheral recruitment of progenitor cells by the endothelium of ischemic tissues. For this purpose, we used HUVEC that we coated with recombinant SDF-1 and then exposed to EPC to promote attachment through CXCR4. In basal conditions, the extent of EPC attachment was found to be consistently higher when issued from the Cav–/– background (+37±13% versus Cav+/+ EPC; P<0.05, n=3) (see Figure 5A and 5B). This adhesion assay also revealed a significantly different pattern when EPC were first exposed to soluble SDF-1 before plating on SDF-1–coated endothelial cells. Indeed, the SDF-1 prechallenge led to a 80% decrease in Cav+/+ EPC adhesion (P<0.01, n=3), whereas the attachment of Cav–/– EPC was not significantly altered (versus unstimulated Cav–/– EPC) (see Figure 5A and 5B). To further confirm the crucial role of caveolin in the adhesiveness of EPC, we transfected EPC from Cav–/– animals with a caveolin-encoding plasmid, and, conversely, EPC from Cav+/+ animals were transduced with caveolin small interfering RNA (siRNA). Caveolin recombinant expression in Cav–/–EPC reached the level of endogenous caveolin in Cav+/+ EPC, whereas after siRNA treatment, caveolin amounted to 15% of the endogenous expression in Cav+/+ EPC (see Figure 5C). Importantly, these acute modifications of Cav+/+ and Cav–/– EPC phenotypes recapitulated the observations made with EPC issued from Cav–/– and Cav+/+ mice, respectively. Accordingly, the reduction in caveolin expression led to a greater adhesion of SDF-1–treated wild-type EPC, and the reexpression of caveolin in Cav–/– EPC dramatically reduced their adhesion capacity in response to SDF-1 (Figure 5C).
Transient Decrease in Caveolin Expression Increases the Vasculogenic Potential of EPC
Finally, we reasoned that the infusion of EPC transiently deprived of caveolin/caveolae should improve their recruitment capacity in ischemic tissues and thereby lead to therapeutic advantages (versus nonmodified EPC). To examine this hypothesis, we treated 1-week cultured EPC with caveolin siRNA to decrease caveolin abundance as described above: the resulting caveolin downregulation (see Figure 5C) was maximal 24 hours after transfection and maintained for the next 36 hours (not shown). To avoid any confounding defects in angiogenesis in Cav–/– mice (versus postnatal vasculogenesis), we induced ischemic hindlimb by femoral artery/vein resection in wild-type animals and examined the ability of EPC treated or not with caveolin siRNA to rescue the limb perfusion. In these wild-type animals, the ischemia-induced neovascularization was very efficient in the adductor muscle around the ligated vessels and hampered the search for biological differences induced by either transduced or nonmodified EPC. We therefore focused on the perfusion and tissue integrity distal from the thigh ligature (eg, in the foot) in the weeks following femoral vessel resection. We found that by infusing caveolin siRNA-transduced EPC, blood flow in the foot from the operated limb was restored to a larger extent than with native EPC infusion (Figure 6A). Similarly, the alterations in both the tissue integrity and an index of the active use of the foot, evaluated by arbitrary scoring, were more extensively corrected with caveolin siRNA-transduced EPC than with sham-transfected EPC (see Figure 6B and 6C).
Discussion
Although many reports using transgenic animals have revealed that the nature of the bone marrow stroma may dramatically influence the progenitor cell mobilization step, the current study emphasizes that the EPC phenotype itself may directly influence their ability to be mobilized. Accordingly, we showed that a defect in caveolin expression could prevent SDF-1–induced internalization of the CXCR4 receptors and the consecutive release of EPC from the bone marrow to the blood circulation. The previously reported defect in postischemic angiogenesis in caveolin-deficient mice,7 therefore, appears to be aggravated by a defect in postnatal vasculogenesis, further highlighting the potential impact of changes in caveolin abundance observed in cardiovascular diseases.20 Furthermore, we found that by reproducing this phenomenon (eg, caveolin downregulation) on isolated, cultured EPC through a caveolin siRNA-based strategy, the SDF-1–dependent homing of progenitor cells to endothelial cells was stimulated and the contribution of postnatal vasculogenesis increased in a model of ischemic tissue reperfusion. These apparently paradoxical data underscore 2 sides of the same coin: caveolae are needed for SDF-1–induced CXCR4 internalization and, although blocking this process in the bone marrow hampers the mobilization of progenitor cells, the transient inhibition of CXCR4 sequestration in isolated, cultured EPC increases their homing capacity in ischemic tissues.
The cell surface expression of the CXCR4 chemokine receptor is a major determinant in the homing capacity of hematopoietic stem and progenitor cells. There are several processes proposed to regulate the SDF-1–mediated mobilization of CXCR4-expressing cells from the bone marrow, including peptidase processing,24,25 ubiquitination and lysosomal sorting,26 and desensitization23,27 of the CXCR4 receptor. In the current study, we identified the SDF-1–stimulated internalization of CXCR4 receptors within caveolae as a critical mode of mobilization from the bone marrow. Indeed, FACS analysis revealed a net decrease in the extent of surface CXCR4 receptors in Cav+/+ but not Cav–/– EPC following soluble SDF-1 exposure. We further provided evidence that the disruption of the CXCR4 direct interaction with SDF-1–expressing stromal cells was involved and not the signaling pathways potentially regulated by caveolin. Previous reports28,29 have shown that signaling downstream SDF-1 is associated with CXCR4 receptor redistribution to rafts (at the leading edge of migrating cells), eventually followed by their internalization. In our study, when isolating detergent-insoluble membranes (essentially made of rafts and caveolae) from the rest of the cell extracts, we found a similar shift of SDF-1–induced CXCR4 receptors toward these detergent-insoluble fractions from both Cav+/+ and Cav–/– EPC (Figure 3D). In Cav+/+ EPC, the detergent-insoluble CXCR4 was further associated to caveolin (as determined by immunoprecipitation), indicating that internalization but not clustering of CXCR4 was altered in Cav–/– EPC. We further documented that in experiments aiming to evaluate the chemotactic potential of SDF-1, no difference could be found between the migratory potential of Cav+/+ and Cav–/– EPC (see supplemental Figure II). Altogether, our data indicate that the caveolin deficiency does not impact the SDF-1–mediated signaling but alters the endocytosis-based dissociation of CXCR4-expressing EPC from SDF-1–decorated stromal cells.
Other groups have identified the formation of clathrin-coated pits as the pathway of CXCR4 receptor internalization.26,30,31 In these studies, however, the authors were using either non–primary transfected cells31 or cells known not to express caveolin/caveolae such as lymphoid cells.30,31 In a more recent study, however, Dar et al reported that SDF-1 could be internalized through clathrin-coated pits and resecreted by bone marrow endothelial and stromal cells.32 Although the clathrin-dependent CXCR4 transcytosis was convincingly documented in this study, the authors did not provide evidence to exclude other routes (such as the caveolae fission), and, more importantly, they reported that for leukocytes and hematopoietic stem cells (that are more parented to the EPC used in our study), clathrin-coated pits were not involved for the transcytosis of SDF-1. Finally, it is worth mentioning that in another situation involving bone marrow mobilization, namely exposure to the cytotoxic 5-FU chemotherapeutic drug, a significant delay in bone marrow reconstitution was observed in Cav–/– mice (not shown), thereby extending the critical role of caveolin/caveolae in the process of BMC trafficking. In this context, the elevated basal levels of plasma SDF-1 in Cav–/– mice (see Figure 4C) may also be interpreted as a compensatory response to a general deficit in HSC mobilization.
Finally, our study reveals that the engagement of progenitor cells in the endothelial lineage is strikingly initiated in the bone marrow. Caveolin expression, as needed for the caveolae sequestration of activated CXCR4 sequestration, is, indeed, induced in the bone marrow on induction of peripheral ischemia. Quantitative evaluation of caveolin mRNA expression in wild-type animals revealed that the abundance of caveolin in hematopoietic cells progressively increased after femoral artery/vein resection. Interestingly, culture of EPC in "endothelial cell growth medium" also revealed a progressive increase in caveolin abundance, suggesting a shift in the phenotype of these cells and their propensity to engage in a "more differentiated" endothelial lineage. Of potential therapeutic interest is the observation that transduction of wild-type EPC with caveolin siRNA to transiently alter caveolae formation allows to exploit a time frame during which progenitor cells are particularly prone to adhere to SDF-1–presenting compartments but after which caveolin-dependent signaling pathways may still occur.
In summary, the present work demonstrates for the first time a critical role of the structural protein caveolin for the EPC mobilization from the bone marrow and the consecutive revascularization of ischemic tissues. The SDF-1 binding to CXCR-4 is, indeed, finely regulated by the caveolar sequestration process, allowing the detachment of progenitor cells from bone marrow stromal cells in the presence of an excess soluble SDF-1 (originating from ischemic tissues). Besides the elucidation of this mechanism that finely regulates the egression of matured progenitor cells, our data also pave the way for the therapeutic exploitation of caveolin targeting (ie, downregulation through siRNA use) to increase the engraftment potential of progenitor and hematopoietic stem cells.
Acknowledgments
This work was supported by grants from the Belgian National Fund for Scientific Research (FNRS), the Belgian Federation against Cancer, the J. Maisin Foundation, an Action de Recherche Concertée (ARC 04/09-317) from the Communauté Franaise de Belgique, and a Ple d’Attraction Interuniversitaire (PAI). O.F. and C.D. are FNRS Research Associates.
Footnotes
Original received September 28, 2005; resubmission received February 16, 2006; accepted March 27, 2006.
References
Isner JM, Asahara T. Angiogenesis and vasculogenesis as therapeutic strategies for postnatal neovascularization. J Clin Invest. 1999; 103: 1231–1236.
Losordo DW, Dimmeler S. Therapeutic angiogenesis and vasculogenesis for ischemic disease. Part I: angiogenic cytokines. Circulation. 2004; 109: 2487–2491.
Asahara T, Kawamoto A. Endothelial progenitor cells for postnatal vasculogenesis. Am J Physiol Cell Physiol. 2004; 287: C572–C579.
Carmeliet P. Angiogenesis in health and disease. Nat Med. 2003; 9: 653–660. [Order article via Infotrieve]
Rafii S, Lyden D. Therapeutic stem and progenitor cell transplantation for organ vascularization and regeneration. Nat Med. 2003; 9: 702–712. [Order article via Infotrieve]
Urbich C, Dimmeler S. Endothelial progenitor cells: characterization and role in vascular biology. Circ Res. 2004; 95: 343–353.
Sonveaux P, Martinive P, DeWever J, Batova Z, Daneau G, Pelat M, Ghisdal P, Gregoire V, Dessy C, Balligand JL, Feron O. Caveolin-1 expression is critical for vascular endothelial growth factor-induced ischemic hindlimb collateralization and nitric oxide-mediated angiogenesis. Circ Res. 2004; 95: 154–161.
Brouet A, DeWever J, Martinive P, Havaux X, Bouzin C, Sonveaux P, Feron O. Antitumor effects of in vivo caveolin gene delivery are associated with the inhibition of the proangiogenic and vasodilatory effects of nitric oxide. FASEB J. 2005; 19: 602–604.
Bauer PM, Yu J, Chen Y, Hickey R, Bernatchez PN, Looft-Wilson R, Huang Y, Giordano F, Stan RV, Sessa WC. Endothelial-specific expression of caveolin-1 impairs microvascular permeability and angiogenesis. Proc Natl Acad Sci U S A. 2005; 102: 204–209.
Gratton JP, Lin MI, Yu J, Weiss ED, Jiang ZL, Fairchild TA, Iwakiri Y, Groszmann R, Claffey KP, Cheng YC, Sessa WC. Selective inhibition of tumor microvascular permeability by cavtratin blocks tumor progression in mice. Cancer Cell. 2003; 4: 31–39. [Order article via Infotrieve]
Gu Z, Kaul M, Yan B, Kridel SJ, Cui J, Strongin A, Smith JW, Liddington RC, Lipton SA. S-nitrosylation of matrix metalloproteinases: signaling pathway to neuronal cell death. Science. 2002; 297: 1186–1190.
Feron O, Kelly RA. The caveolar paradox: suppressing, inducing, and terminating eNOS signaling. Circ Res. 2001; 88: 129–131.
Wei Y, Yang X, Liu Q, Wilkins JA, Chapman HA. A role for caveolin and the urokinase receptor in integrin-mediated adhesion and signaling. J Cell Biol. 1999; 144: 1285–1294.
Frank PG, Lee H, Park DS, Tandon NN, Scherer PE, Lisanti MP. Genetic ablation of caveolin-1 confers protection against atherosclerosis. Arterioscler Thromb Vasc Biol. 2004; 24: 98–105.
Upla P, Marjomaki V, Kankaanpaa P, Ivaska J, Hyypia T, Van Der Goot FG, Heino J. Clustering induces a lateral redistribution of alpha 2 beta 1 integrin from membrane rafts to caveolae and subsequent protein kinase C-dependent internalization. Mol Biol Cell. 2004; 15: 625–636.
Ge S, Pachter JS. Caveolin-1 knockdown by small interfering RNA suppresses responses to the chemokine monocyte chemoattractant protein-1 by human astrocytes. J Biol Chem. 2004; 279: 6688–6695.
Razani B, Woodman SE, Lisanti MP. Caveolae: from cell biology to animal physiology. Pharmacol Rev. 2002; 54: 431–467.
Parton RG, Richards AA. Lipid rafts and caveolae as portals for endocytosis: new insights and common mechanisms. Traffic. 2003; 4: 724–738. [Order article via Infotrieve]
Feron O, Dessy C, Moniotte S, Desager JP, Balligand JL. Hypercholesterolemia decreases nitric oxide production by promoting the interaction of caveolin and endothelial nitric oxide synthase. J Clin Invest. 1999; 103: 897–905.
Sbaa E, Frerart F, Feron O. The double regulation of endothelial nitric oxide synthase by caveolae and caveolin: a paradox solved through the study of angiogenesis. Trends Cardiovasc Med. 2005; 15: 157–162. [Order article via Infotrieve]
Drab M, Verkade P, Elger M, Kasper M, Lohn M, Lauterbach B, Menne J, Lindschau C, Mende F, Luft FC, Schedl A, Haller H, Kurzchalia TV. Loss of caveolae, vascular dysfunction, and pulmonary defects in caveolin-1 gene-disrupted mice. Science. 2001; 293: 2449–2452.
Vasa M, Fichtlscherer S, Adler K, Aicher A, Martin H, Zeiher AM, Dimmeler S. Increase in circulating endothelial progenitor cells by statin therapy in patients with stable coronary artery disease. Circulation. 2001; 103: 2885–2890. [Order article via Infotrieve]
Ceradini DJ, Kulkarni AR, Callaghan MJ, Tepper OM, Bastidas N, Kleinman ME, Capla JM, Galiano RD, Levine JP, Gurtner GC. Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med. 2004; 10: 858–864. [Order article via Infotrieve]
Sadir R, Imberty A, Baleux F, Lortat-Jacob H. Heparan sulfate/heparin oligosaccharides protect stromal cell-derived factor-1 (SDF-1)/CXCL12 against proteolysis induced by CD26/dipeptidyl peptidase IV. J Biol Chem. 2004; 279: 43854–43860.
Valenzuela-Fernandez A, Planchenault T, Baleux F, Staropoli I, Le Barillec K, Leduc D, Delaunay T, Lazarini F, Virelizier JL, Chignard M, Pidard D, Arenzana-Seisdedos F. Leukocyte elastase negatively regulates stromal cell-derived factor-1 (SDF-1)/CXCR4 binding and functions by amino-terminal processing of SDF-1 and CXCR4. J Biol Chem. 2002; 277: 15677–15689.
Marchese A, Benovic JL. Agonist-promoted ubiquitination of the G protein-coupled receptor CXCR4 mediates lysosomal sorting. J Biol Chem. 2001; 276: 45509–45512.
Shen H, Cheng T, Olszak I, Garcia-Zepeda E, Lu Z, Herrmann S, Fallon R, Luster AD, Scadden DT. CXCR-4 desensitization is associated with tissue localization of hemopoietic progenitor cells. J Immunol. 2001; 166: 5027–5033.
van Buul JD, Voermans C, van Gelderen J, Anthony EC, van der Schoot CE, Hordijk PL. Leukocyte-endothelium interaction promotes SDF-1-dependent polarization of CXCR4. J Biol Chem. 2003; 278: 30302–30310.
Nguyen DH, Taub D. CXCR4 function requires membrane cholesterol: implications for HIV infection. J Immunol. 2002; 168: 4121–4126.
Signoret N, Oldridge J, Pelchen-Matthews A, Klasse PJ, Tran T, Brass LF, Rosenkilde MM, Schwartz TW, Holmes W, Dallas W, Luther MA, Wells TN, Hoxie JA, Marsh M. Phorbol esters and SDF-1 induce rapid endocytosis and down modulation of the chemokine receptor CXCR4. J Cell Biol. 1997; 139: 651–664.
Venkatesan S, Rose JJ, Lodge R, Murphy PM, Foley JF. Distinct mechanisms of agonist-induced endocytosis for human chemokine receptors CCR5 and CXCR4. Mol Biol Cell. 2003; 14: 3305–3324.
Dar A, Goichberg P, Shinder V, Kalinkovich A, Kollet O, Netzer N, Margalit R, Zsak M, Nagler A, Hardan I, Resnick I, Rot A, Lapidot T. Chemokine receptor CXCR4-dependent internalization and resecretion of functional chemokine SDF-1 by bone marrow endothelial and stromal cells. Nat Immunol. 2006; 6: 1038–1046.
the Unit of Pharmacology and Therapeutics, University of Louvain Medical School, Brussels, Belgium.
Abstract
When neovascularization is triggered in ischemic tissues, angiogenesis but also (postnatal) vasculogenesis is induced, the latter requiring the mobilization of endothelial progenitor cells (EPC) from the bone marrow. Caveolin, the structural protein of caveolae, was recently reported to directly influence the angiogenic process through the regulation of the vascular endothelial growth factor (VEGF)/nitric oxide pathway. In this study, using caveolin-1 null mice (Cav–/–), we examined whether caveolin was also involved in the EPC recruitment in a model of ischemic hindlimb. Intravenous infusion of Sca-1+ Lin– progenitor cells, but not bone marrow transplantation, rescued the defective neovascularization in Cav–/– mice, suggesting a defect in progenitor mobilization. The adhesion of Cav–/– EPC to bone marrow stromal cells indeed appeared to be resistant to the otherwise mobilizing SDF-1 (Stromal cell–Derived Factor-1) exposure because of a defect in the internalization of the SDF-1 cognate receptor CXCR4. Symmetrically, the attachment of Cav–/– EPC to SDF-1–presenting endothelial cells was significantly increased. Finally, EPC transduction with caveolin small interfering RNA reproduced this advantage in vitro and, importantly, led to a more extensive rescue of the ischemic hindlimb after intravenous infusion (versus sham-transfected EPC). These results underline the critical role of caveolin in ensuring the caveolae-mediated endocytosis of CXCR4, regulating both the SDF-1–mediated mobilization and peripheral homing of progenitor cells in response to ischemia. In particular, a transient reduction in caveolin expression was shown to therapeutically increase the engraftment of progenitor cells.
Key Words: caveolin caveolae postnatal vasculogenesis endothelial progenitor cells ischemia
Introduction
Vasculogenesis and angiogenesis are the 2 major processes contributing to neovascularization in adults.1,2 Postnatal vasculogenesis in adults requires the recruitment of progenitor cells derived from the bone marrow,3 whereas angiogenesis involves the proliferation and migration of endothelial cells from the preexisting vessels.4 The proportion of either process in promoting the development of neovessels in pathological situations, such as ischemia and tumor growth, may vary and both are considered to be complementary.3,5 Moreover, the 2 processes share biological mediators including chemocytokines and proteinases.6
Among the key players in postnatal vasculogenesis, SDF-1 (Stromal cell–Derived Factor-1) and vascular endothelial growth factor (VEGF) have undoubtedly been the focus of the most attention. Through specific interactions with the CXCR4 and VEGFR-2 receptors, SDF-1 and VEGF, respectively, contribute to increase the recruitment of progenitor cells from the bone marrow to the peripheral blood, thereby leading to the formation of new vessels at the ischemic sites.2 Recently, we identified a critical role of caveolin-1 (a structural protein of caveolae) in the regulation of the VEGF-signaling pathway in a model of adaptive angiogenesis.7 We found that the compartmentation of the VEGFR-2 receptor in caveolae was key to ensure VEGF-mediated neovascularization. Accordingly, genetic deletion of caveolin-1 (and the ensuing absence of caveolae) was found to alter the endothelial nitric oxide synthase (eNOS) and ERK-dependent pathways downstream of VEGF stimulation, thereby leading to a defect in angiogenesis. A key role of caveolin-1 in angiogenesis was also identified in mouse tumor models and in the implanted Matrigel plug assay,8–10 mostly through the modulation of nitric oxide bioavailability. Notably, in none of these studies focusing on the role of caveolin in neovascularization was the potential impact of caveolin/caveolae in postnatal vasculogenesis (versus angiogenesis) addressed.
A series of indirect observations, however, argue for a role of caveolin in the bone marrow cell–dependent neovessel formation: (1) the activity of matrix metalloprotease-9 (MMP-9), a major protease involved in stem cell mobilization is modulated by nitric oxide,11 the production of which is directly dependent on caveolin abundance12; (2) some proteins involved in the interaction between progenitor cells and bone marrow stroma, including adhesion molecules13–15 and chemokine receptor,16 have already been associated to caveolae in other tissues; (3) the quasilinear relationship between caveolin expression and caveolae density,17 and the capacity of caveolae to pinch off and to sequester their content18 may account for the dynamic regulation of the expression of 1 or several of the abovementioned proteins at the cell surface.
Such evidence suggesting the potential role of caveolin/caveolae in progenitor cell biology and mobilization prompted us to explore the vasculogenic process in caveolin-deficient mice. More generally, another purpose of this study was to determine whether promotion of postnatal vasculogenesis could help rescue ischemic tissues in caveolin-null mice. Alteration in caveolin abundance has, indeed, been reported in many cardiovascular diseases that themselves are associated with a defect in angiogenesis.19,20
Materials and Methods
An expanded Materials and Methods section is provided in the online data supplement at http://circres.ahajournals.org.
Animals
Caveolin-deficient mice (Cav–/–)21 and their control littermates (Cav+/+) were generated through heterozygous matings and housed in our local facility. Anesthetized mice underwent a double femoral artery and vein ligation to allow resection of 2-mm vessel segments under a stereoscopic microscope while innervation was carefully preserved.
Bone Marrow and Endothelial Progenitor Cells
Bone marrow mononuclear cells were isolated by separation onto a Ficoll Histopaque gradient (Sigma), and magnetic cell sorting (Miltenyi Biotech) was used for positive and/or negative selection to collect Sca-1+/lineage marker-negative (Lin–) cells or CD45+ cells (BD Bioscience Pharmingen). For endothelial progenitor cell (EPC) production, bone marrow mononuclear cells were directly cultured on fibronectin-coated flasks using endothelial cell growth medium, as previously described.22 The identity of EPC was confirmed by morphological and immunocytochemical analyses. EPC were used at day 7 after initial seeding except when indicated otherwise.
Results
Infusion of Sca-1+ Lin– Bone Marrow Cells Rescues Ischemic Hindlimb Perfusion in Cav–/– Mice
We have previously reported that after resection of 2 mm-femoral artery/vein segments, Cav–/– mice never restored a normal hindlimb perfusion, compared with wild-type mice that completely recovered7 (see also below). To evaluate whether progenitor cells could help rescuing the impaired neovascularization in Cav–/– mice, we examined the effects of the infusion of Sca-1+ Lin– BMC on the restoration of the perfusion in the operated hindlimb. Serial examinations of blood flow by laser Doppler imaging were performed on days 3, 7, 10, and 14 (day 0 being fixed as the day of ligature). Figure 1A shows that infusion of 106 BMC issued either from wild-type (Cav+/+) or Cav–/– mice did rescue the limb perfusion, whereas a permanent drop in blood flow was observed in the operated leg of untreated Cav–/– mice. Of note, in Cav+/+ mice, infusion of Cav+/+ BMC barely influenced the extent (and rate) of the otherwise naturally occurring reperfusion of the operated hindlimb (see Figure 1B).
In some experiments, we loaded BMC with a long-lasting tracker (carboxymethyl-1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine [CM-DiI]) to follow their in vivo fate after infusion. Accordingly, histological analyses confirmed the presence of CM-DiI–labeled cells in the adductor muscle of Cav–/– mice, regardless of the Cav–/– or Cav+/+ mouse origin of infused cells (Figure 1C). CM-DiI–labeled cells could be found as early as 24 hours after induction of ischemia and were still detectable at day 14 (at much lower level). It should, however, be noted that posttransplantation dilution of the tracker (with cell division) may lead to underestimation of progenitor cell recruitment at longer times. The CM-DiI overlapping with CD31-positive capillary endothelial cells amounted to 29±12% and 37±13% (of total CM-DiI staining) in experiments with Cav–/– and Cav+/+ BMC, respectively (Figure 1C).
Ischemia-Induced Mobilization of Cav–/– Progenitor Cells Is Defective
Because infusion of Cav–/– BMC in Cav–/– mice did restore normal blood flow, a defect in mobilization could be suspected in these mice. We, therefore, set up a series of experiments where bone marrow CD45+ cells (eg, depleted of stromal cells) from Cav+/+ and Cav–/– mice were transplanted into lethally irradiated Cav–/– mice; Cav+/+ mice were similarly transplanted to allow full comparison between the different combinations of donor and recipient mice. A 3-week delay was respected after transplantation to ensure the complete reconstitution of bone marrow; preliminary studies had shown that circulating leukocyte number and bone marrow integrity were back to normal 14 days after CD45+ cell infusion. Figure 2A, left panels, shows that Cav–/– mice recovered from the ischemic insult when CD45+ cells from wild-type mice, but not from Cav–/– mice, were used to reconstitute the bone marrow. In Cav+/+ background animals (Figure 2A, right panels), the reconstitution of bone marrow also led to distinct patterns of response according to the origin of CD45+ cells, although to a lesser extent than in Cav–/– animals. Accordingly, when Cav–/– CD45+ cells were used, the restoration of blood flow in the proximal operated limb was delayed when compared with wild-type CD45+-reconstituted mice (see Figure 2A, right panels). In fact, whereas in the latter mice, no distal limb atrophy was detectable, the loss of toes or the entire foot of the operated leg was consistently observed in Cav–/– CD45+-reconstituted wild-type mice. These morphological alterations were less prominent, however, than in Cav–/– CD45+-reconstituted Cav–/– mice; a complete leg atrophy occurred in a majority of these mice (5 of 6).
Based on the above results, besides the transplanted mouse background, the origin of bone marrow seemed also critical for postischemic revascularization (eg, no or delayed recovery with CD45+ cells originated from Cav–/– mice). We therefore repeated the above experiments by using CM-DiI–labeled CD45+ cells to reconstitute the bone marrow of irradiated mice (see Figure I in the online data supplement) and to further track the progenitor cell homing in the ischemic hindlimb. In Cav–/– mice (Figure 2B, left panels), progenitor cells derived from the Cav–/– CD45+ cell transplantation were found heterogeneously distributed across the ischemic muscle and, in total, much less abundant than those arising from Cav+/+ cell transplantation (see Figure 2C for quantitative analysis). In Cav+/+ mice, although similar extents of CM-DiI–labeled progenitor cells issued either from Cav+/+ or Cav–/– mice could be found in some regions of the adductor muscle (see Figure 2B, right panels), the distribution was less homogeneous in the latter condition. Accordingly, quantitative analyses of the all tissue sections revealed a lesser staining when progenitor cells were issued from Cav–/– CD45+ cell transplantation (see bars 3 and 4 in Figure 2C).
The microvascular density (see CD31 immunolabeling in Figure 2B, lower panels) was significantly higher in both mouse strains when the CD45+ donor mouse was from a wild-type background (see Figure 2D for quantitative analysis); colocalization of CM-DiI and CD31 was however very poor suggesting paracrine influence of progenitor cells instead of direct participation to the structure of new capillaries. Moreover, slight but statistically significant differences distinguish the animals from Cav+/+ and Cav–/– backgrounds: (1) the extent of CM-DiI–labeled Cav+/+ progenitor cells found in the ischemic muscle was larger in Cav–/– recipient (compare bars 2 and 4 in Figure 2C); (2) the extent of CD31–labeled cells was larger in Cav+/+ background animals (compare bars 1 and 3 in Figure 2D).
Cav–/– EPC Are Resistant to SDF-1–Induced Detachment From Bone Marrow Stromal Cells
To further dissect the mechanism accounting for the defect in mobilization of Cav–/– BMC, we examined whether the MMP-9 and SDF-1 pathways were differently regulated in wild-type and caveolin-deficient mice. Zymography analyses of crude bone marrow extracts did not reveal significant differences in gelatinase activity (not shown), suggesting the absence of alterations in MMP-2 and -9 activities in Cav–/– mice. Furthermore, because we were anticipating alterations in the EPC phenotype more than in the bone marrow stroma (where most of MMP activity originates), we focused our attention on SDF-1/CXCL12 and its receptor CXCR4 located on EPC. Accordingly, we isolated stromal cells from Cav–/– and Cav+/+ by negative selection (CD45–) from total bone marrow and cultured them to confluence in the presence of serum. Stromal cell monolayers were then exposed to hypoxia (1% O2 for 4 hours) to promote endogenous SDF-1 expression as previously documented.23 Adhesion experiments were performed with EPC originating from both mouse genotypes, prechallenged or not with SDF-1. Figure 3A shows that when EPC were issued from Cav+/+ mice, the SDF-1 treatment led to a decreased ability to adhere to stromal cells independently of their genotype (–85±7% and –72±10% with stromal cells from Cav+/+ and Cav–/–, respectively, versus untreated EPC; P<0.01, n=3). Conversely, Cav–/– EPC were resistant to SDF-1 stimulation and did not reveal significant reduction in adhesion when compared with untreated EPC (see also Figure 3B for quantitative analysis). In the same experimental set up, however, CXCR4 antibody blockade could abolish the Cav+/+ and Cav–/– EPC adhesion on Cav–/– stromal cells to the same extent, further confirming the implication of the SDF-1/CXCR4 axis in the caveolin-mediated mobilization of EPC (Figure 3B). Fluorescence-activated cell sorting (FACS) analysis was then performed with EPC exposed or not to SDF-1 to examine the level of cell surface expression of the CXCR4 receptor, the SDF-1 cognate receptor. Although the total amounts of CXCR4 receptors were not significantly different in untreated Cav+/+ and Cav–/– EPC, the SDF-1 treatment did induce a shift to the left of the fluorescent peak in Cav+/+ EPC, indicating the disappearance of part of CXCR4 receptors from the cell surface (Figure 3C). Such shift in fluorescence was not observed with SDF-1–treated Cav–/– EPC, thereby illustrating the resistance to internalization of CXCR4 when caveolin was absent (Figure 3C). We also performed CXCR4 immunoblotting on detergent-insoluble membranes enriched in rafts and caveolae, from both Cav+/+ and Cav–/– EPC (Figure 3D). We found that in EPC from both origins, SDF-1 similarly induced the translocation of a large proportion of CXCR4 receptors to the detergent-insoluble fractions (approximately +40% of total CXCR4 receptors; see the shift between lanes 1 and 2 in Figure 3D). The implication of clathrin-coated pits was further excluded because the use of the cationic amphiphilic drug chlorpromazine, an inhibitor of clathrin-dependent, receptor-mediated endocytosis, failed to affect the CXCR4 internalization (Figure 3D). More importantly, when caveolin immunoprecipitation was performed to remove caveolar membranes (eg, to discriminate from rafts), the CXCR4 receptor could not be found in the remaining supernatant of Cav+/+ EPC membranes, confirming the key role of caveolae in CXCR4 sequestration (see lane 4 in Figure 3D).
Ischemia Leads to an Increase in BMC Caveolin Expression and in Plasma SDF-1 Levels
Because the data presented above indicated a critical role of caveolin and SDF-1 in EPC mobilization, we examined how their expression evolved with time in situ. Accordingly, we used real time PCR to first compare the extent of caveolin-1 mRNA transcripts in Sca-1+ Lin– BMC collected at different time points after induction of hindlimb ischemia. Figure 4A shows that although caveolin-1 is almost absent from these hematopoietic cells in normal conditions, its expression rapidly increased in response to peripheral ischemia to reach a maximum after 7 days. The elevated level of caveolin expression in Sca-1+ Lin– BMC was maintained for 2 more weeks before rapidly dropping back to basal values. Finally, we examined whether cultured EPC also exhibited changes in caveolin abundance following in vitro exposure to serum and VEGF. Figure 4B shows that, indeed, the level of caveolin progressively increased to reach a maximum at day 14 of culture. Also, evaluation of circulating levels of SDF-1 revealed that in basal conditions, plasma levels of SDF-1 were 2-fold higher in Cav–/– than in Cav+/+ mice. Following ischemic insult, the SDF-1 plasma levels increased in Cav+/+ mice to reach the level of Cav–/– mice (Figure 4C). Interestingly, however, the amounts of circulating EPC (determined by numeration of colonies) appeared to be 41±19% higher in Cav+/+ mouse plasma (versus Cav+/+ mice; P<0.01, n=7).
Lack of Caveolin Increases the EPC Adhesion to Endothelial Cells
We next examined whether alterations in the caveolin-mediated CXCR4 internalization, which translated in a defect in mobilization from the bone marrow, also impacted the peripheral recruitment of progenitor cells by the endothelium of ischemic tissues. For this purpose, we used HUVEC that we coated with recombinant SDF-1 and then exposed to EPC to promote attachment through CXCR4. In basal conditions, the extent of EPC attachment was found to be consistently higher when issued from the Cav–/– background (+37±13% versus Cav+/+ EPC; P<0.05, n=3) (see Figure 5A and 5B). This adhesion assay also revealed a significantly different pattern when EPC were first exposed to soluble SDF-1 before plating on SDF-1–coated endothelial cells. Indeed, the SDF-1 prechallenge led to a 80% decrease in Cav+/+ EPC adhesion (P<0.01, n=3), whereas the attachment of Cav–/– EPC was not significantly altered (versus unstimulated Cav–/– EPC) (see Figure 5A and 5B). To further confirm the crucial role of caveolin in the adhesiveness of EPC, we transfected EPC from Cav–/– animals with a caveolin-encoding plasmid, and, conversely, EPC from Cav+/+ animals were transduced with caveolin small interfering RNA (siRNA). Caveolin recombinant expression in Cav–/–EPC reached the level of endogenous caveolin in Cav+/+ EPC, whereas after siRNA treatment, caveolin amounted to 15% of the endogenous expression in Cav+/+ EPC (see Figure 5C). Importantly, these acute modifications of Cav+/+ and Cav–/– EPC phenotypes recapitulated the observations made with EPC issued from Cav–/– and Cav+/+ mice, respectively. Accordingly, the reduction in caveolin expression led to a greater adhesion of SDF-1–treated wild-type EPC, and the reexpression of caveolin in Cav–/– EPC dramatically reduced their adhesion capacity in response to SDF-1 (Figure 5C).
Transient Decrease in Caveolin Expression Increases the Vasculogenic Potential of EPC
Finally, we reasoned that the infusion of EPC transiently deprived of caveolin/caveolae should improve their recruitment capacity in ischemic tissues and thereby lead to therapeutic advantages (versus nonmodified EPC). To examine this hypothesis, we treated 1-week cultured EPC with caveolin siRNA to decrease caveolin abundance as described above: the resulting caveolin downregulation (see Figure 5C) was maximal 24 hours after transfection and maintained for the next 36 hours (not shown). To avoid any confounding defects in angiogenesis in Cav–/– mice (versus postnatal vasculogenesis), we induced ischemic hindlimb by femoral artery/vein resection in wild-type animals and examined the ability of EPC treated or not with caveolin siRNA to rescue the limb perfusion. In these wild-type animals, the ischemia-induced neovascularization was very efficient in the adductor muscle around the ligated vessels and hampered the search for biological differences induced by either transduced or nonmodified EPC. We therefore focused on the perfusion and tissue integrity distal from the thigh ligature (eg, in the foot) in the weeks following femoral vessel resection. We found that by infusing caveolin siRNA-transduced EPC, blood flow in the foot from the operated limb was restored to a larger extent than with native EPC infusion (Figure 6A). Similarly, the alterations in both the tissue integrity and an index of the active use of the foot, evaluated by arbitrary scoring, were more extensively corrected with caveolin siRNA-transduced EPC than with sham-transfected EPC (see Figure 6B and 6C).
Discussion
Although many reports using transgenic animals have revealed that the nature of the bone marrow stroma may dramatically influence the progenitor cell mobilization step, the current study emphasizes that the EPC phenotype itself may directly influence their ability to be mobilized. Accordingly, we showed that a defect in caveolin expression could prevent SDF-1–induced internalization of the CXCR4 receptors and the consecutive release of EPC from the bone marrow to the blood circulation. The previously reported defect in postischemic angiogenesis in caveolin-deficient mice,7 therefore, appears to be aggravated by a defect in postnatal vasculogenesis, further highlighting the potential impact of changes in caveolin abundance observed in cardiovascular diseases.20 Furthermore, we found that by reproducing this phenomenon (eg, caveolin downregulation) on isolated, cultured EPC through a caveolin siRNA-based strategy, the SDF-1–dependent homing of progenitor cells to endothelial cells was stimulated and the contribution of postnatal vasculogenesis increased in a model of ischemic tissue reperfusion. These apparently paradoxical data underscore 2 sides of the same coin: caveolae are needed for SDF-1–induced CXCR4 internalization and, although blocking this process in the bone marrow hampers the mobilization of progenitor cells, the transient inhibition of CXCR4 sequestration in isolated, cultured EPC increases their homing capacity in ischemic tissues.
The cell surface expression of the CXCR4 chemokine receptor is a major determinant in the homing capacity of hematopoietic stem and progenitor cells. There are several processes proposed to regulate the SDF-1–mediated mobilization of CXCR4-expressing cells from the bone marrow, including peptidase processing,24,25 ubiquitination and lysosomal sorting,26 and desensitization23,27 of the CXCR4 receptor. In the current study, we identified the SDF-1–stimulated internalization of CXCR4 receptors within caveolae as a critical mode of mobilization from the bone marrow. Indeed, FACS analysis revealed a net decrease in the extent of surface CXCR4 receptors in Cav+/+ but not Cav–/– EPC following soluble SDF-1 exposure. We further provided evidence that the disruption of the CXCR4 direct interaction with SDF-1–expressing stromal cells was involved and not the signaling pathways potentially regulated by caveolin. Previous reports28,29 have shown that signaling downstream SDF-1 is associated with CXCR4 receptor redistribution to rafts (at the leading edge of migrating cells), eventually followed by their internalization. In our study, when isolating detergent-insoluble membranes (essentially made of rafts and caveolae) from the rest of the cell extracts, we found a similar shift of SDF-1–induced CXCR4 receptors toward these detergent-insoluble fractions from both Cav+/+ and Cav–/– EPC (Figure 3D). In Cav+/+ EPC, the detergent-insoluble CXCR4 was further associated to caveolin (as determined by immunoprecipitation), indicating that internalization but not clustering of CXCR4 was altered in Cav–/– EPC. We further documented that in experiments aiming to evaluate the chemotactic potential of SDF-1, no difference could be found between the migratory potential of Cav+/+ and Cav–/– EPC (see supplemental Figure II). Altogether, our data indicate that the caveolin deficiency does not impact the SDF-1–mediated signaling but alters the endocytosis-based dissociation of CXCR4-expressing EPC from SDF-1–decorated stromal cells.
Other groups have identified the formation of clathrin-coated pits as the pathway of CXCR4 receptor internalization.26,30,31 In these studies, however, the authors were using either non–primary transfected cells31 or cells known not to express caveolin/caveolae such as lymphoid cells.30,31 In a more recent study, however, Dar et al reported that SDF-1 could be internalized through clathrin-coated pits and resecreted by bone marrow endothelial and stromal cells.32 Although the clathrin-dependent CXCR4 transcytosis was convincingly documented in this study, the authors did not provide evidence to exclude other routes (such as the caveolae fission), and, more importantly, they reported that for leukocytes and hematopoietic stem cells (that are more parented to the EPC used in our study), clathrin-coated pits were not involved for the transcytosis of SDF-1. Finally, it is worth mentioning that in another situation involving bone marrow mobilization, namely exposure to the cytotoxic 5-FU chemotherapeutic drug, a significant delay in bone marrow reconstitution was observed in Cav–/– mice (not shown), thereby extending the critical role of caveolin/caveolae in the process of BMC trafficking. In this context, the elevated basal levels of plasma SDF-1 in Cav–/– mice (see Figure 4C) may also be interpreted as a compensatory response to a general deficit in HSC mobilization.
Finally, our study reveals that the engagement of progenitor cells in the endothelial lineage is strikingly initiated in the bone marrow. Caveolin expression, as needed for the caveolae sequestration of activated CXCR4 sequestration, is, indeed, induced in the bone marrow on induction of peripheral ischemia. Quantitative evaluation of caveolin mRNA expression in wild-type animals revealed that the abundance of caveolin in hematopoietic cells progressively increased after femoral artery/vein resection. Interestingly, culture of EPC in "endothelial cell growth medium" also revealed a progressive increase in caveolin abundance, suggesting a shift in the phenotype of these cells and their propensity to engage in a "more differentiated" endothelial lineage. Of potential therapeutic interest is the observation that transduction of wild-type EPC with caveolin siRNA to transiently alter caveolae formation allows to exploit a time frame during which progenitor cells are particularly prone to adhere to SDF-1–presenting compartments but after which caveolin-dependent signaling pathways may still occur.
In summary, the present work demonstrates for the first time a critical role of the structural protein caveolin for the EPC mobilization from the bone marrow and the consecutive revascularization of ischemic tissues. The SDF-1 binding to CXCR-4 is, indeed, finely regulated by the caveolar sequestration process, allowing the detachment of progenitor cells from bone marrow stromal cells in the presence of an excess soluble SDF-1 (originating from ischemic tissues). Besides the elucidation of this mechanism that finely regulates the egression of matured progenitor cells, our data also pave the way for the therapeutic exploitation of caveolin targeting (ie, downregulation through siRNA use) to increase the engraftment potential of progenitor and hematopoietic stem cells.
Acknowledgments
This work was supported by grants from the Belgian National Fund for Scientific Research (FNRS), the Belgian Federation against Cancer, the J. Maisin Foundation, an Action de Recherche Concertée (ARC 04/09-317) from the Communauté Franaise de Belgique, and a Ple d’Attraction Interuniversitaire (PAI). O.F. and C.D. are FNRS Research Associates.
Footnotes
Original received September 28, 2005; resubmission received February 16, 2006; accepted March 27, 2006.
References
Isner JM, Asahara T. Angiogenesis and vasculogenesis as therapeutic strategies for postnatal neovascularization. J Clin Invest. 1999; 103: 1231–1236.
Losordo DW, Dimmeler S. Therapeutic angiogenesis and vasculogenesis for ischemic disease. Part I: angiogenic cytokines. Circulation. 2004; 109: 2487–2491.
Asahara T, Kawamoto A. Endothelial progenitor cells for postnatal vasculogenesis. Am J Physiol Cell Physiol. 2004; 287: C572–C579.
Carmeliet P. Angiogenesis in health and disease. Nat Med. 2003; 9: 653–660. [Order article via Infotrieve]
Rafii S, Lyden D. Therapeutic stem and progenitor cell transplantation for organ vascularization and regeneration. Nat Med. 2003; 9: 702–712. [Order article via Infotrieve]
Urbich C, Dimmeler S. Endothelial progenitor cells: characterization and role in vascular biology. Circ Res. 2004; 95: 343–353.
Sonveaux P, Martinive P, DeWever J, Batova Z, Daneau G, Pelat M, Ghisdal P, Gregoire V, Dessy C, Balligand JL, Feron O. Caveolin-1 expression is critical for vascular endothelial growth factor-induced ischemic hindlimb collateralization and nitric oxide-mediated angiogenesis. Circ Res. 2004; 95: 154–161.
Brouet A, DeWever J, Martinive P, Havaux X, Bouzin C, Sonveaux P, Feron O. Antitumor effects of in vivo caveolin gene delivery are associated with the inhibition of the proangiogenic and vasodilatory effects of nitric oxide. FASEB J. 2005; 19: 602–604.
Bauer PM, Yu J, Chen Y, Hickey R, Bernatchez PN, Looft-Wilson R, Huang Y, Giordano F, Stan RV, Sessa WC. Endothelial-specific expression of caveolin-1 impairs microvascular permeability and angiogenesis. Proc Natl Acad Sci U S A. 2005; 102: 204–209.
Gratton JP, Lin MI, Yu J, Weiss ED, Jiang ZL, Fairchild TA, Iwakiri Y, Groszmann R, Claffey KP, Cheng YC, Sessa WC. Selective inhibition of tumor microvascular permeability by cavtratin blocks tumor progression in mice. Cancer Cell. 2003; 4: 31–39. [Order article via Infotrieve]
Gu Z, Kaul M, Yan B, Kridel SJ, Cui J, Strongin A, Smith JW, Liddington RC, Lipton SA. S-nitrosylation of matrix metalloproteinases: signaling pathway to neuronal cell death. Science. 2002; 297: 1186–1190.
Feron O, Kelly RA. The caveolar paradox: suppressing, inducing, and terminating eNOS signaling. Circ Res. 2001; 88: 129–131.
Wei Y, Yang X, Liu Q, Wilkins JA, Chapman HA. A role for caveolin and the urokinase receptor in integrin-mediated adhesion and signaling. J Cell Biol. 1999; 144: 1285–1294.
Frank PG, Lee H, Park DS, Tandon NN, Scherer PE, Lisanti MP. Genetic ablation of caveolin-1 confers protection against atherosclerosis. Arterioscler Thromb Vasc Biol. 2004; 24: 98–105.
Upla P, Marjomaki V, Kankaanpaa P, Ivaska J, Hyypia T, Van Der Goot FG, Heino J. Clustering induces a lateral redistribution of alpha 2 beta 1 integrin from membrane rafts to caveolae and subsequent protein kinase C-dependent internalization. Mol Biol Cell. 2004; 15: 625–636.
Ge S, Pachter JS. Caveolin-1 knockdown by small interfering RNA suppresses responses to the chemokine monocyte chemoattractant protein-1 by human astrocytes. J Biol Chem. 2004; 279: 6688–6695.
Razani B, Woodman SE, Lisanti MP. Caveolae: from cell biology to animal physiology. Pharmacol Rev. 2002; 54: 431–467.
Parton RG, Richards AA. Lipid rafts and caveolae as portals for endocytosis: new insights and common mechanisms. Traffic. 2003; 4: 724–738. [Order article via Infotrieve]
Feron O, Dessy C, Moniotte S, Desager JP, Balligand JL. Hypercholesterolemia decreases nitric oxide production by promoting the interaction of caveolin and endothelial nitric oxide synthase. J Clin Invest. 1999; 103: 897–905.
Sbaa E, Frerart F, Feron O. The double regulation of endothelial nitric oxide synthase by caveolae and caveolin: a paradox solved through the study of angiogenesis. Trends Cardiovasc Med. 2005; 15: 157–162. [Order article via Infotrieve]
Drab M, Verkade P, Elger M, Kasper M, Lohn M, Lauterbach B, Menne J, Lindschau C, Mende F, Luft FC, Schedl A, Haller H, Kurzchalia TV. Loss of caveolae, vascular dysfunction, and pulmonary defects in caveolin-1 gene-disrupted mice. Science. 2001; 293: 2449–2452.
Vasa M, Fichtlscherer S, Adler K, Aicher A, Martin H, Zeiher AM, Dimmeler S. Increase in circulating endothelial progenitor cells by statin therapy in patients with stable coronary artery disease. Circulation. 2001; 103: 2885–2890. [Order article via Infotrieve]
Ceradini DJ, Kulkarni AR, Callaghan MJ, Tepper OM, Bastidas N, Kleinman ME, Capla JM, Galiano RD, Levine JP, Gurtner GC. Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med. 2004; 10: 858–864. [Order article via Infotrieve]
Sadir R, Imberty A, Baleux F, Lortat-Jacob H. Heparan sulfate/heparin oligosaccharides protect stromal cell-derived factor-1 (SDF-1)/CXCL12 against proteolysis induced by CD26/dipeptidyl peptidase IV. J Biol Chem. 2004; 279: 43854–43860.
Valenzuela-Fernandez A, Planchenault T, Baleux F, Staropoli I, Le Barillec K, Leduc D, Delaunay T, Lazarini F, Virelizier JL, Chignard M, Pidard D, Arenzana-Seisdedos F. Leukocyte elastase negatively regulates stromal cell-derived factor-1 (SDF-1)/CXCR4 binding and functions by amino-terminal processing of SDF-1 and CXCR4. J Biol Chem. 2002; 277: 15677–15689.
Marchese A, Benovic JL. Agonist-promoted ubiquitination of the G protein-coupled receptor CXCR4 mediates lysosomal sorting. J Biol Chem. 2001; 276: 45509–45512.
Shen H, Cheng T, Olszak I, Garcia-Zepeda E, Lu Z, Herrmann S, Fallon R, Luster AD, Scadden DT. CXCR-4 desensitization is associated with tissue localization of hemopoietic progenitor cells. J Immunol. 2001; 166: 5027–5033.
van Buul JD, Voermans C, van Gelderen J, Anthony EC, van der Schoot CE, Hordijk PL. Leukocyte-endothelium interaction promotes SDF-1-dependent polarization of CXCR4. J Biol Chem. 2003; 278: 30302–30310.
Nguyen DH, Taub D. CXCR4 function requires membrane cholesterol: implications for HIV infection. J Immunol. 2002; 168: 4121–4126.
Signoret N, Oldridge J, Pelchen-Matthews A, Klasse PJ, Tran T, Brass LF, Rosenkilde MM, Schwartz TW, Holmes W, Dallas W, Luther MA, Wells TN, Hoxie JA, Marsh M. Phorbol esters and SDF-1 induce rapid endocytosis and down modulation of the chemokine receptor CXCR4. J Cell Biol. 1997; 139: 651–664.
Venkatesan S, Rose JJ, Lodge R, Murphy PM, Foley JF. Distinct mechanisms of agonist-induced endocytosis for human chemokine receptors CCR5 and CXCR4. Mol Biol Cell. 2003; 14: 3305–3324.
Dar A, Goichberg P, Shinder V, Kalinkovich A, Kollet O, Netzer N, Margalit R, Zsak M, Nagler A, Hardan I, Resnick I, Rot A, Lapidot T. Chemokine receptor CXCR4-dependent internalization and resecretion of functional chemokine SDF-1 by bone marrow endothelial and stromal cells. Nat Immunol. 2006; 6: 1038–1046.
您现在查看是摘要介绍页，详见ORG附件。