OUP user menu

CXCR4 gene transfer contributes to in vivo reendothelialization capacity of endothelial progenitor cells

Long Chen, Fang Wu, Wen-hao Xia, Yuan-yuan Zhang, Shi-yue Xu, Fei Cheng, Xin Liu, Xiao-yu Zhang, Shen-ming Wang, Jun Tao
DOI: http://dx.doi.org/10.1093/cvr/cvq207 462-470 First published online: 23 June 2010

Abstract

Aims Endothelial progenitor cells (EPCs) play a pivotal role in endothelial repair after artery injury. The chemokine receptor CXCR4 is a key modulator of the homing of EPCs to impaired artery and reendothelialization. In this study, we addressed the hypothesis that CXCR4 gene transfer could enhance the reendothelialization capacity of EPCs.

Methods and results In vitro, human EPCs were expanded and transduced with adenovirus serotype 5 encoding the human CXCR4 gene (Ad5/CXCR4). In vitro, CXCR4 gene transfer augmented EPC migration and enhanced EPC adhesion to endothelial cell monolayers. Adhesion assays under flow conditions showed that CXCR4 gene transfer increased the ability of EPCs to arrest on fibronectin. To determine whether CXCR4 gene transfer facilitated therapeutic reendothelialization, the effect of EPCs on in vivo reendothelialization was examined in nude mice subjected to carotid artery injury. Compared with the vehicle, transplantation of EPCs with or without gene transfer significantly accelerated in vivo reendothelialization; however, transplantation of EPCs transduced with Ad5/CXCR4 had a further enhanced effect compared with control EPCs containing EPCs transduced with an adenovirus encoding enhanced green fluorescent protein gene or non-transduced EPCs. We also found that phosphorylation of Janus kinase-2 (JAK-2), a CXCR4 downstream signalling target, was increased in EPCs transduced with Ad5/CXCR4. The enhanced in vitro function and in vivo reendothelialization capacity of EPCs by CXCR4 gene transfer were abolished by neutralizing antibodies against CXCR4 or/and JAK-2 inhibitor AG490.

Conclusion The present study demonstrates that CXCR4 gene transfer contributes to the enhanced in vivo reendothelialization capacity of EPCs. Up-regulation of CXCR4 in human EPCs may become a novel therapeutic target for endothelial repair.

  • Endothelium
  • Reendothelialization
  • Endothelial progenitor cells
  • CXCR4
  • Gene therapy

1. Introduction

It is generally accepted that loss of the normal integrity of endothelial structure and function is involved in the initiation and development of atherosclerotic vascular disease.14 Accelerated endothelial restoration is an important therapeutic means for vascular repair in injured artery.5,6 Accumulating evidence indicates that circulating endothelial progenitor cells (EPCs) derived from the bone marrow play a pivotal role in the homeostasis of damaged vascular repair by enhanced reendothelialization.79 Transplantation of EPCs onto the balloon-injured arteries10 or prosthetic vascular grafts11,12 led to formation of a bioactive endothelial monolayer associated with significant inhibition of neointimal formation and thrombosis. However, impaired EPC function including mobilization from the bone marrow and homing to vascular lesion was present in several cardiovascular risk factors such as ageing, diabetes, hypertension, and hyperlipidaemia, which limits the beneficial effect of EPCs on reendothelialization.1315 EPC gene transfer during in vitro expansion constitutes a potential means to improve such limitations in EPC function and may be a novel gene- and cell-based therapeutic strategy to prevent and treat the atherosclerotic vascular disease.

It has been demonstrated that CXCR4, receptor of chemokine stromal cell-derived factor-1 (SDF-1), is a key molecule in regulating EPC homing16 and CXCR4 signalling blockade contributes to the reduced homing of EPCs to injured artery.17 Accordingly, CXCR4 may be an available molecular target of gene therapy for EPCs. Until now, there are no data to show the impact of CXCR4 gene transfer on in vitro functional properties and in vivo reendothelialization capacity of EPCs. Given the close association between CXCR4 signalling and EPCs, we hypothesized that CXCR4 gene transfer might promote EPC-mediated endothelial repair after artery injury. To address this assumption, we evaluated the effect of CXCR4 gene transfer on in vitro migration and adhesion of human EPCs as well as the efficacy of CXCR4 gene transfer pre-treatment as a strategy to enhance in vivo reendothelialization after wire-mediated injury of carotid artery in nude mice. We also investigated CXCR4-mediated Janus kinase-2 (JAK-2) signalling related to both the in vitro function and the in vivo reendothelialization capacity of EPCs.

2. Methods

2.1 EPC culture, materials, and cell labelling

EPCs were isolated and cultured as previously described in detail.12,1821 In brief, peripheral blood mononuclear cells from healthy subjects (aged from 25 to 35 years old) were isolated by Ficoll density gradient centrifugation and were cultured on fibronectin-coated six-well plates in endothelial cell basal medium-2 (EBM-2) supplemented with endothelial growth medium SingleQuots exactly as indicated by the manufacturer (Clonetics, San Diego, CA, USA). After 4 days culture, non-adherent cells were removed by washing plates with phosphate-buffered solution (PBS), and new medium was applied. Adherent cells were maintained for 7 days and then were used for the following experiments. Neutralizing monoclonal antibody against CXCR4 (CXCR4-mAb), IgG2a isotype control (R&D System, Minneapolis, MN, USA), and JAK-2 inhibitor AG490 (Alexis, Plymouth Meeting, PA, USA) were used as blocking agents in this study.

EPCs were defined as cells dually positive for acLDL uptake and lectin binding. Cultured EPCs were incubated with DiI-acLDL (0.02 mg/mL; Invitrogen, Carlsbad, CA, USA) for 2 h in a cell incubator. Subsequently, cells were washed and fixed with 4% (v/v) paraformaldehyde for 15 min and incubated with FITC-labelled BS-1 lectin (0.01 mg/mL; Sigma-Aldrich, St Louis, MO, USA) for 1 h. Plates of cells were again washed and incubated with a DAPI nuclear counterstain. Double-positive cells were observed with a fluorescent microscope (×200 magnification; Olympus, Tokyo, Japan).

2.2 Flow cytometry analysis

Endothelial marker proteins were examined by flow cytometry analysis using phycoerythrin (PE)-labelled monoclonal mouse anti-human antibodies recognizing CD31 (BD, San Diego, CA, USA), von Willebrand factor (vWF), and kinase-insert domain receptor (KDR) (R&D System). Furthermore, expression of the monocytic lineage marker CD14 (BD) was analysed. Surface CXCR4 expression was detected by PE-labelled monoclonal mouse anti-human antibodies (BD). To determine the expression of each of these surface antigens, cells were incubated for 40 min at 4°C in a volume of 100 µL with an appropriate amount of PE-labelled antibody or corresponding IgG isotype control. At least 1 × 105 EPCs were acquired using a flow cytometry (Beckman–Coulter, Fullerton, CA, USA).

2.3 EPC gene transfer

After 7 days in culture, cells were transduced with the adenovirus serotype 5 (Ad5) encoding the human CXCR4 gene (Ad5/CXCR4) or enhanced green fluorescent protein gene (Ad5/EGFP) (Vector Gene Technology Company Ltd, Beijing, China). To establish optimum virus concentration for EPC adenovirus gene transfer, different multiplicities of infection were evaluated as indicated by the adenovirus manufacturer. After preliminary experiments were performed, human EPCs were transduced with 1000 MOI Ad5/CXCR4 or Ad/EGFP for 90 min in culture medium without serum. After transduction, cells were washed with PBS and incubated with EPC medium for 48 h before subsequent experiments.

2.4 RT–PCR and western blot analysis

Total RNA was isolated with the mRNA abstraction kit (Tiangen Biotech, Beijing, China). RT–PCR was carried out by the routine two-step method. The primer CXCR4 A (sense) is 5′-TCTTCCTGCCCACCATCTACTC-3′, and the primer CXCR4 B (antisense) is 5′-GTAGATGACATGGACTGCCTTGC-3′.

EPC protein was harvested by cell lysis buffer (Cell Signaling Technology, Boston, MA, USA). Protein extracts were subjected to SDS–PAGE, transferred to polyvinylidene fluoride membranes (Roche, Indianapolis, IN, USA). The following antibodies were used: rabbit anti-CXCR4 antibody (1:1000; abCAM, Cambridge, MA, USA), rabbit anti-phospho-JAK-2 and anti-JAK-2 antibody (1:1000; Cell Signaling Technology), and rabbit anti-actin antibody (1:2000; Cell Signaling Technology). Proteins were visualized with HRP-conjugated anti-rabbit IgG (1:2000; Cell Signaling Technology), followed by use of the ECL chemiluminescence system (Cell Signaling Technology). To detect SDF-1-stimulated phosphorylation of JAK-2, EPCs were pre-incubated with 100 ng/mL SDF-1 (Peprotech, Rocky Hill, NJ, USA) for 10 min before protein harvesting.

2.5 EPC adhesion to endothelial cells in vitro

A monolayer of human umbilical vein endothelial cells (HUVECs) was prepared 48 h before the assay by plating 2 × 105 cells in each well of a four-well plate. HUVECs were pre-treated with or without 1 ng/mL tumour necrosis factor-α (TNF-α, Peprotech) for 12 h. Then 1 × 105 CM-DiI (CellTracker™ CM-DiI, Invitrogen)-labelled EPCs were added to each well and incubated for 3 h at 37°C. Non-attached cells were gently removed with PBS, and adherent EPCs were fixed with 4% paraformaldehyde and counted by independent investigators blinded to treatment randomly.

2.6 EPC migration in vitro

A total of 2 × 104 EPCs were isolated, resuspended in 250 µL EBM-2, and pipetted at the seventh day in the upper chamber of a modified Boyden chamber (Costar Transwell® assay, 8 µm pore size, Corning, NY, USA). The chamber was placed in a 24-well culture dish containing 500 µL EBM-2 supplemented with either PBS or 100 ng/mL SDF-1. After 24 h incubation at 37°C, transmigrated cells were counted by independent investigators blinded to treatment randomly.

2.7 In vitro EPC adhesion assays in flow

Laminar flow assays were performed as Hristov et al.17 described previously. Dishes were coated with fibronectin (10 µg/mL), and EPCs were stimulated with or without 100 ng/mL SDF-1 for 10 min before the assays. EPCs (5 × 105/mL) were then resuspended in assay buffer (HEPES-buffered Hank's balanced salt solution, 1 mmol/L Mg2+/Ca2+, and 0.5% BSA) and perfused into the flow chamber (proprietary item; commercially obtained from RWTH Aachen University) at a shear rate of 1.5 dyn/cm2 for 4 min at 37°C. The number of adherent cells after 4 min was quantified in multiple fields by independent investigators blinded to treatment randomly.

2.8 Animals and in vivo reendothelialization assay

Male NRMInu/nu athymic nude mice (SLAC Laboratory Animal Center, Shanghai, China), aged 8–10 weeks, were used to allow injection of human EPCs. Animals were anaesthetized with ketamine (100 mg/kg ip) and xylazine (5 mg/kg ip). Surgery was carried out using a dissecting microscope. The left carotid artery was exposed via a midline incision on the ventral side of the neck. The bifurcation of the carotid artery was located, and two ligatures were placed around the external carotid artery, which was then tied off with the distal ligature. An incision hole was made between the two ligatures to introduce the denudation device. The curved flexible wire (0.35 mm diameter) was introduced into the common carotid artery and passed three times in order to denude endothelium. The wire was then removed, and the external carotid artery was tied off proximal to the incision hole with the proximal ligature.

EPCs (5 × 105 cells) were resuspended in 100 µL of pre-warmed PBS (37°C) and transplanted 3 h after carotid artery injury via tail vein injection with a 27 G needle. The same volume of PBS was injected into placebo mice. Three days after carotid artery injury, endothelial regeneration was evaluated by staining denuded areas with 100 µL of solution containing 3% Evans Blue dye via tail vein injection. A fluorescent microscope (Olympus) was performed to detect homing of transplanted EPCs to the site of vascular injury in separate experiments (n = 5) with the use of CM-DiI-labelled EPCs.

All experimental protocols complied with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and the Animal Care and Use Committees of Sun Yat-sen University. Our study conformed to the ethical principles outlined in the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Sun Yat-sen University (Guangzhou, China).

2.9 Statistical analysis

All results are expressed as mean ± SEM. Statistical significance was evaluated by means of Student's t-test or ANOVA. A value of P < 0.05 was considered to denote statistical significance. All statistical analyses used SPSS statistical software (SPSS version 13.0).

3. Results

3.1 EPC characterization

Recently, two distinct populations of EPCs have been described according to their time-dependent appearance in culture, that is, early outgrowth EPCs appearing after 4–7 days, and late outgrowth EPCs appearing after 14–21 days.22,23 A beneficial effect on endothelial repair after injury has been shown exactly for early outgrowth EPCs.17,24,25

In this study, EPCs were cultivated out of peripheral blood mononuclear cells from healthy subjects on fibronectin-coated dishes for 7 days. Fluorescent photos showed that vast majority of adherent cells were double-positive for acLDL–lectin staining (Figure 1A), meanwhile expressed endothelial marker proteins (CD31, vWF, and KDR) and a monocytic marker CD14 at comparable levels (Figure 1B). All the characterization above indicated that the cultured EPCs in this study could be classified into early outgrowth EPCs as previously.2226

Figure 1

Characterization of cultured EPCs. (A) Representative photographs of EPCs at the seventh day labelled with DAPI (blue), DiI-acLDL (red), FITC-lectin (green), and merged image (×200 magnification, scale bar = 100 µm). (B) Flow cytometry analysis of the endothelial markers CD31, vWF, KDR, and the monocytic lineage marker CD14 of EPCs (IgG isotype control shown in red, n = 5 per group). Numbers are the mean ± SEM percentage of positive cells for all experiments determined by comparison with corresponding negative control labelling.

3.2 Ad5/CXCR4 transfection up-regulated CXCR4 expression of EPCs

Human CXCR4 cDNA was cloned into the Ad5 vector. The transcription and expression of CXCR4 in EPCs were confirmed by RT–PCR and western blot. The CXCR4 mRNA and protein level were obviously up-regulated in Ad5/CXCR4-transduced EPCs compared with Ad5/EGFP-transduced EPCs or non-transduced EPCs (Figure 2A and B). Furthermore, the flow cytometry analysis showed that the level of surface CXCR4 was obviously higher in Ad5/CXCR4-transduced EPCs than in Ad5/EGFP-transduced EPCs or non-transduced EPCs (Figure 2C).

Figure 2

Ad5/CXCR4 transfection up-regulated CXCR4 expression of EPCs. (A) The CXCR4 mRNA level in non-Adv-EPCs, Ad5/EGFP-EPCs, and Ad5/CXCR4-EPCs was measured by RT–PCR, and actin was used for control (n = 3 per group); *P < 0.05 vs. non-Adv-EPCs or Ad5/EGFP-EPCs. (B) The expression of CXCR4 protein in non-Adv-EPCs, Ad5/EGFP-EPCs, and Ad5/CXCR4-EPCs was measured by western blot, and actin was used for control (n = 3 per group); *P < 0.05 vs. non-Adv-EPCs or Ad5/EGFP-EPCs. (C) The flow cytometry analysis for surface CXCR4 expression on EPCs (IgG isotype control shown in red, n = 3 per group). Representative images and histogram plots; *P < 0.05 vs. non Adv-EPCs or Ad5/EGFP-EPCs.

3.3 CXCR4 gene transfer improved adhesion and migration of EPCs in vitro

At 48 h after transduction, EPCs were labelled with the fluorescent marker CM-DiI for cell tracking. CM-DiI-labelled EPCs were incubated on an HUVEC monolayer with or without TNF-α (1 ng/mL) pre-treatment for 12 h. After 3 h of incubation, non-adherent cells were removed by washing with PBS gently, and CM-DiI-labelled cells adherent to the HUVEC monolayer were manually counted (Figure 3A). In the quiescent HUVEC monolayer, adhesion of CM-DiI-labelled EPCs was not significantly different among non-transduced EPCs, Ad5/EGFP-transduced EPCs, and Ad5/CXCR4-transduced EPCs. However, in TNF-α-activated HUVECs, adhesion of Ad5/CXCR4-transduced EPCs exceeded Ad5/EGFP-transduced EPCs or non-transduced EPCs (65 ± 8.9 vs. 39 ± 2.8 or 41 ± 3.5; Figure 3B).

Figure 3

Adhesion and migration function assays of EPCs in vitro. (A) Representative photographs of non-Adv-EPCs, Ad5/EGFP-EPCs, and Ad5/CXCR4-EPCs adherent to HUVEC monolayer with or without TNF-α activation. EPCs labelled with CM-DiI (red) and all cells stained with DAPI (blue) for counting (×200 magnification, scale bar = 100 µm). (B) Quantification analysis of EPC adhesion to HUVECs with or without TNF-α activation (n = 5 per group); *P < 0.05 vs. non-Adv-EPCs + TNF-α or Ad5/EGFP-EPCs + TNF-α. (C) Quantification analysis of basal migration and SDF-1 induced migration of non-Adv-EPCs, Ad5/EGFP-EPCs, and Ad5/CXCR4-EPCs (n = 5 per group); *P < 0.05 vs. non-Adv-EPCs + SDF-1 or Ad5/EGFP-EPCs + SDF-1. (D) Representative photographs of non Adv-EPCs, Ad5/EGFP-EPCs, and Ad5/CXCR4-EPCs adherent to fibronectin with or without SDF-1 stimulation. Arrested EPCs stained with DAPI (blue) for counting (×200 magnification, scale bar = 100 µm). (E) Quantification analysis of EPC adhesion to fibronectin with or without SDF-1 stimulation (n = 5 per group); *P < 0.05 vs. non Adv-EPCs + SDF-1 or Ad5/EGFP-EPCs + SDF-1.

In the migration assay, the basal migration capacity had no obvious difference among non-transduced EPCs, Ad5/EGFP-transduced EPCs, and Ad5/CXCR4-transduced EPCs. Ad5/CXCR4-transduced EPCs had a markedly enhanced migration response to SDF-1 compared with Ad5/EGFP-transduced EPCs or non-transduced EPCs (83 ± 8.1 vs. 58 ± 4.6 or 56 ± 4.9; Figure 3C).

Using the flow chamber, we analysed the adhesion of EPCs to fibronectin in flow. The number of adhesion was not significantly different among non-transduced EPCs, Ad5/EGFP-transduced EPCs, and Ad5/CXCR4-transduced EPCs without SDF-1 stimulation. After stimulation with SDF-1, the adhesion of EPCs from all the three groups was increased, and the Ad5/CXCR4-transduced EPCs had an obviously enhanced response compared with the other two groups (92 ± 11.0 vs. 70 ± 8.8 or 68 ± 9.2; Figure 3D and E).

3.4 CXCR4 gene transfer enhanced in vivo reendothelialization capacity of EPCs

The carotid endothelium injury was confirmed by Evans Blue staining (Figure 4A). Fluorescent microscope analysis of a subgroup of nude mice (n = 5) revealed that transplanted EPCs attached at the site of vascular injury (Figure 5).

Figure 4

Reendothelialization of injured carotid arteries is promoted by EPC transplantation and further enhanced by CXCR4 gene transfer. (A) Endothelium injury was confirmed by Evans Blue dye staining immediately after wire-mediated carotid artery injury in nude mice. Representative photographs: injured artery and contralateral uninjured artery (arteries without discission). (B) Reendothelialized area at day 3 after carotid injury in nude mice with PBS injection (n = 6), transplantation of non-Adv-EPCs (n = 5), Ad5/EGFP-EPCs (n = 5), or Ad5/CXCR4-EPCs (n = 6). *P < 0.01 vs. PBS group, **P < 0.05 vs. non-Adv-EPCs or Ad5/EGFP-EPCs group; and representative photographs (arteries were cut parallel to long axis of the carotid).

Figure 5

EPC tracking in vivo. (A) Representative photographs under fluorescent microscope: (top four panels) carotid artery 3 days after injury showing CM-DiI-labelled EPCs (red) attached to FITC-lectin-stained endothelium (green), nuclei stained with DAPI (blue), for higher magnification, see (B); (bottom four panels) contralateral uninjured carotid artery. (A) ×200 magnification; (B) ×400 magnification, scale bar = 100 µm.

Werner et al.27 first described cultured EPCs in vitro enhanced arterial repair in vivo. We hypothesized that the enhanced function in vitro achieved by CXCR4 gene transfer would translate into improved function in vivo. To test this hypothesis, PBS, non-transduced EPCs, Ad5/EGFP-transduced EPCs, and Ad5/CXCR4-transduced EPCs were systemically injected into nude mice subjected to carotid artery injury. Notably, compared with the PBS, treatment with non-transduced EPCs, Ad5/EGFP-transduced EPCs, or Ad5/CXCR4-transduced EPCs obviously increased reendothelialization of denuded carotid arteries in nude mice, but Ad5/CXCR4-transduced EPCs had further enhanced effect (61.5 ± 2.4 vs. 39.2 ± 4.9 or 40.5 ± 3.7 vs. 7.4 ± 1.1%; Figure 4B).

3.5 CXCR4-mediated JAK-2 signalling was associated with reendothelialization capacity of EPCs

Because JAK-2 is a known downstream target of the CXCR4 receptor,28 we investigated whether JAK-2 activity was increased by CXCR4 gene transfer in EPCs. Immunoblotting revealed that the basal JAK-2-phosphorylation was not significantly different between Ad5/EGFP-transduced EPCs and Ad5/CXCR4-transduced EPCs; however, SDF-1-stimulated increase in JAK-2 phosphorylation in Ad5/CXCR4-transduced EPCs exceeded that in Ad5/EGFP-transduced EPCs. Furthermore, SDF-1-stimulated increase in JAK-2 phosphorylation in Ad5/CXCR4-transduced EPCs was inhibited by 10 µg/mL CXCR4-mAb or 10 µmol/L AG490 pre-incubation for 30 min (Figure 6A).

Figure 6

JAK-2 signalling related to the enhanced in vitro function and in vivo reendothelialization capacity of EPCs by CXCR4 gene transfer. (A) Representative JAK-2 phosphorylation of EPCs with or without stimulation by SDF-1 was measured by western blot (n = 3 per group); quantification expressed as p-JAK-2/JAK-2 ratio. *P < 0.05 vs. Ad5/EGFP-EPCs + SDF-1; #P < 0.05 vs. Ad5/CXCR4-EPCs + SDF-1. (B) Quantification analysis for adhesion of EPCs to TNF-α-activated HUVECs after treatment with CXCR4-mAb and/or AG490 (n = 5 per group). *P < 0.01 vs. Ad5/EGFP-EPCs baseline; #P < 0.01 vs. Ad5/CXCR4-EPCs baseline. (C) Quantification analysis for SDF-1-induced migration of EPCs after treatment with CXCR4-mAb and/or AG490 (n = 5 per group). *P < 0.01 vs. Ad5/EGFP-EPCs baseline; #P < 0.01 vs. Ad5/CXCR4-EPCs baseline. (D) Quantification analysis for adhesion of EPCs to fibronectin in flow after treatment with CXCR4-mAb and/or AG490 (n = 5 per group). *P < 0.01 vs. Ad5/EGFP-EPCs baseline; #P < 0.01 vs. Ad5/CXCR4-EPCs baseline. (E) Reendothelialized area at day 3 after carotid injury in nude mice with transplantation of Ad5/CXCR4-EPCs (n = 6), CXCR4-mAb-treated Ad5/CXCR4-EPCs (n = 5), or AG490-treated Ad5/CXCR4-EPCs (n = 5); *P < 0.05 vs. Ad5/CXCR4-EPCs group.

We also investigated whether CXCR4-mediated JAK-2 signalling was associated with enhanced reendothelialization capacity of EPCs with CXCR4 gene transfer. Our data showed that pre-incubation with either CXCR4-mAb or AG490 profoundly reduced in vitro functional properties (endothelial adhesion, migration, and adhesion in flow) of Ad5/EGFP-EPCs and Ad5/CXCR4-EPCs, meanwhile combined JAK-2 and CXCR4 blockade did not result in an additive effect (Figure 6BD). Consistent with the results of in vitro function assays, pre-treatment with CXCR4-mAb or AG490 can also attenuate the further enhanced in vivo reendothelialization capacity of Ad5/CXCR4-transduced EPCs (Figure 6E). In these assays, we used IgG2a isotype antibody as the control of CXCR4-mAb. Isotype control antibody was ineffective in blocking adhesion and migratory activity and inhibiting JAK-2 phosphorylation of EPCs (data not shown).

4. Discussion

Endothelial integrity is critical for the homeostasis of vascular structure and function.14 Accelerating reendothelialization following artery injury is very important for vascular repair.5,6 Maintenance of normal number and function of EPCs in systemic circulation is now known as an important novel endogenous vascular repair factor.79 It has been shown that after vascular injury, EPCs are quickly recruited by cytokines and inflammatory factors to seed the new intimal lining on the injured region29 and stimulate the neighbouring endothelial cells migration and proliferation by secreting angiogenic growth factors,30 contributing to accelerated reendothelialization. Therefore, increasing EPC number and function may facilitate arterial repair—particularly, the restoration of the endothelium. Recent studies have proposed that up-regulation of functional properties of EPCs is a novel strategy to enhance EPC-mediated reendothelialization. Among them, the capacity of EPCs homing to impaired artery is increasingly highlighted.3133 Various molecules may be involved in the process of EPC homing. It has been demonstrated that CXCR4 plays a key role in modulating EPC homing.16 Previous investigation showed that stimulation and/or sensitization of CXCR4-mediated signalling can facilitate EPC-mediated neovascularization.34 However, to our knowledge, there is no study to evaluate the effect of CXCR4 gene transfer on EPC-mediated reendothelialization after artery injury. In the current study, we hypothesized that direct up-regulation of CXCR4 expression in human EPCs by gene transfer results in a functional enhancement of EPCs in vitro as well as improvement of reendothelialization capacity in vivo.

To address these assumptions, we first up-regulated CXCR4 in human EPCs by gene transfer with Ad5/CXCR4. Then, we observed the effect of CXCR4 gene transfer on adhesion and migration properties of EPCs. We found that Ad5/CXCR4-transduced EPCs, compared with Ad5/EGFP-transduced or non-transduced EPCs, had a markedly increased adhesion to TNF-α-activated endothelial cells and improved migration response to SDF-1. Furthermore, we used a flow chamber to detect the adhesion profile of EPCs on fibronectin in flow and the data showed that the adhesive properties of EPCs under flow conditions were significantly increased by CXCR4 gene transfer. To test whether the increase in functional properties of EPCs contributes to augmentation of EPC-mediated in vivo reendothelialization, we established the carotid artery-injured nude mice model and investigated the effect of transplantation of EPCs transduced with Ad5/CXCR4 on in vivo reendothelialization. We also found that contrast to Ad5/EGFP-transduced or non-transduced EPCs, transplantation of Ad5/CXCR4-transduced EPCs significantly enlarged reendothelialization area in the carotid artery-injured nude mice model. The salutary effects of CXCR4 gene transfer on both in vitro function and in vivo reendothelialization capacity of EPCs were inhibited by neutralizing antibodies against CXCR4. It has been reported that CXCR4 blockade attenuated EPCs homing to sites of artery injury,17 suggesting that the impaired CXCR4 signalling might be an important molecular mechanism for reduced reendothelialization capacity of EPCs. Our present study, taken with previous investigation,17 demonstrated for the first time that up-regulation of CXCR4 contributes to in vivo reendothelialization capacity of EPCs and CXCR4 is a therapeutic molecular target for augmenting EPC function.

It is well-known that JAK-2 is the important downstream effector molecule of CXCR4 and involves in the CXCR4-mediated cell function alteration.28 In order to understand the molecular mechanism, we hypothesized that JAK-2 signalling might be related to the enhanced EPC function by CXCR4 gene transfer mentioned above. Indeed, the data reported here showed that JAK-2 phosphorylation was more responsive to SDF-1 in Ad5/CXCR4-transduced EPCs compared with Ad5/EGFP-transduced EPCs. Moreover, both the enhanced in vitro functional properties and in vivo reendothelialization capacity of Ad5/CXCR4-transduced EPCs can be attenuated by JAK-2 inhibitor AG490, and combined CXCR4 and JAK-2 blockade did not lead to additive effect. These results suggested that enhanced EPC function by CXCR4 gene transfer was mainly dependent on CXCR4-mediated JAK-2 signal pathway, consistent with previous study.28

Concerning how CXCR4 gene transfer modulates reendothelialization, we are inclined to think that CXCR4 gene transfer augments EPCs homing to sites of artery injury. Hristov et al.17 showed that CXCR4 blockade attenuated EPCs homing to arterial lesion sites. Also, Walter et al.28 reported that CXCR4-blocked EPCs and CXCR4-deficient EPCs displayed diminished migration and invasion function in vitro. In this study, our data suggested that EPCs with CXCR4 gene transfer exhibited enhanced adhesion and migration capacity in vitro. Moreover, these functional alterations of EPCs in vitro were paralleled to increased recruitment of in vivo transplanted EPCs to the sites of vascular injury. Thus, it can be speculated that up-regulating CXCR4 in EPCs by gene transfer enhanced reendothelialization via augmented homing capacity of EPCs to injured arteries.

The present study has a clinical implication to develop a novel therapeutic strategy for subjects with endothelial injury, in whom accelerated reendothelialization is an effective approach to reduce the incidence of atherosclerotic vascular disease. It has been demonstrated that the patients with atherosclerotic vascular disease, who need endothelial repair, are also the same individuals with lack of EPCs.10,11,35 Hence, it is essential to increase the EPC function to compensate the loss in number of EPCs. We supposed that EPCs with high expression of CXCR4 can supply more benefits for endothelial repair. The notion put forward here is further evidenced by our present study, in which transplantation of EPCs with high expression of CXCR4 leads to high reendothelialization efficacy following artery injury.

The present study is subject to several limitations. First, although CXCR4 gene transfer results in accelerated reendothelialization following artery injury, the feasibility of this strategy for the clinical application may be limited due to its inconvenience of transplantation of manipulated cells in vitro. Secondly, the clinical safety of the adenovirus as a gene therapy vector should still be verified. Thirdly, the patients with atherosclerotic vascular disease are associated with disorders of EPC function; however, we did not investigate the effect of CXCR4 gene transfer on dysfunctional EPCs from these subjects. Finally, we just demonstrated the enhanced repair capacity of EPCs with CXCR4 gene transfer in a relative simple in vivo environment—the young and healthy nude mice. Whether this strategy can resist clinical complex environment remains to be elucidated. We should consider these questions described above in our future studies as well.

In summary, the present study demonstrates that up-regulation of CXCR4 by gene transfer improves in vitro functional properties of human EPCs and enhances EPC-mediated in vivo arterial reendothelialization—a novel observation with significant potential for therapeutic development. CXCR4-mediated JAK-2 signalling is, at least in part, involved in this beneficial effect. CXCR4 may hopefully become a novel therapeutic molecular target in clinic endothelial repair and deserve further investigation.

Funding

This work was supported by grants from National Natural Scientific Foundation of China (u0732002, 30973535, and 30770895) and the PhD Programs Foundation of Ministry of Education of China (20090171110061).

Acknowledgements

We thank the Department of Maternity of the First Affiliated Hospital of Sun Yat-Sen University for supplying the human umbilical vein endothelial cells for the present research.

Conflict of interest: none declared.

References

View Abstract