BASIC RESEARCH STUDIES

Transcription

1 BASIC RESEARCH STUDIES Combination of stromal-derived factor-1 and vascular endothelial growth factor gene-modified endothelial progenitor cells is more effective for ischemic neovascularization Jian-Xing Yu, MD, PhD, a Xue-Fei Huang, MD, PhD, b Wei-Ming Lv, MD, PhD, a Cai-Sheng Ye, MD, PhD, a Xin-Zhi Peng, MD, a Hui Zhang, MD, c Long-Bin Xiao, MD, d and Shen-Ming Wang, MD, PhD, a Guangzhou, People s Republic of China Background: Recruitment and entrapment of bone marrow-derived endothelial progenitor cells (EPCs) is important in vascular endothelial growth factor (VEGF)-induced angiogenesis. EPC mobilization and differentiation are modulated by stromal-derived factor-1 (SDF-1 /CXCL12), another important chemokine. In this study, we investigated the hypothesis that SDF-1 and VEGF might act synergistically on EPC-mediated vasculogenesis. Methods: EPCs were isolated and cultured from human peripheral blood, then transduced with retroviral vectors pbabe containing human VEGF 165 complimentary DNA (Td/V-EPCs) and pbabe wild-type (Td/p-EPCs). EPC migration activity was investigated with a modified Boyden chamber assay. EPC apoptosis induced by serum starvation was studied by annexin V assays. The combined effect of local administration of SDF-1 and Td/V-EPC transplantation on neovascularization was investigated in a murine model of hind limb ischemia. Results: Over-expression of hvegf 165 increased SDF-1 mediated EPC migration. SDF-1 mediated migration was significantly increased when EPCs were modified with VEGF (Td/V-EPCs) vs when VEGF was not present (Td/p-EPCs) or when VEGF alone was present (Td/V-EPCs; , , and /mm 2, respectively P <.001). SDF-1 combined with VEGF reduced serum starvation-induced apoptosis of EPCs more than SDF-1 or VEGF alone (P <.001). To determine the effect of this combination in vivo, SDF-1 was locally injected alone into the ischemic hind limb muscle of nude mice or combined with systemically injected Td/V-EPCs. The SDF-1 plus VEGF group showed significantly increased local accumulation of EPCs, blood-flow recovery, and capillary density compared with the other groups. The ratio of ischemic/normal blood flow in Td/V-EPCs plus SDF-1 group was significantly higher (P <.01), as was capillary density (capillaries/mm 2 ), an index of neovascularization (Td/V-EPCs plus SDF-1 group, ; no treatment, ; SDF-1, ; Td/p-EPCs, ; Td/p-EPCs plus SDF-1, ; Td/V-EPCs, ; P <.01). To investigate a possible mechanistic basis, we showed that VEGF up-regulated the receptor for SDF-1, CXCR4, on EPCs in vitro. Conclusion: The combination of SDF-1 and VEGF greatly increases EPC-mediated angiogenesis. The use VEGF and SDF-1 together, rather than alone, will be a novel and efficient angiogenesis strategy to provide therapeutic neovascularization. (J Vasc Surg 2009;50: ) Clinical Relevance: The stimulation of neovascularization using growth factors is a promising experimental treatment for arterial occlusive disease. Combined therapy could be better than the use of single growth factors for enhancing therapeutic angiogenesis, especially in severely ischemic tissue. This study suggests that stromal-derived factor-1 (SDF-1 ) and vascular endothelial growth factor (VEGF) form a synergistic angiogenic pathway that is critical for endothelial progenitor cell-induced neovascularization. Targeting both SDF-1 and VEGF signals may be a novel, efficient strategy for treating ischemic diseases. From the Departments of Vascular Surgery, a Hematology, b and Anesthesiology, c The First Affiliated Hospital, and the Department of Gastrointestinal- Pancreatic Surgery, Huangpu Division of the Hospital, d The First Affiliated Hospital, Sun Yat-Sen University. This work was supported by the grants from the Science and Technology Program of Guangdong Province (2003C32712) and the Natural Science Foundation of Guangdong Province (13092), and the Natural Science Foundation of China ( ). Competition of interest: none. Reprint requests: Shen-Ming Wang, MD, PhD, Department of Vascular Surgery, The First Affiliated Hospital, Sun Yat-Sen University, 58 Zhongshan Road II, Guangzhou , PR China ( wangshenming550@126.com) /$36.00 Copyright 2009 by the Society for Vascular Surgery. doi: /j.jvs Two substances important in neovascularization are stromal cell-derived factor 1 (SDF-1 ) and vascular endothelial growth factor (VEGF). SDF-1 is increased in acutely ischemic tissue, and the SDF gradient formed between the ischemic tissue and the peripheral blood attracts endothelial progenitor cells (EPCs) from blood to the ischemic tissue. 1,2 Although SDF-1 levels are increased in acutely ischemic tissue, they are low in chronically ischemic tissue. 3 When this cytokine is injected into chronically ischemic tissue, it increases EPC recruitment but does not cause cell proliferation. 4 VEGF is a mitogen that does increase EPC proliferation. However, as with SDF-1 levels, tissue VEGF levels are decreased in chronic ischemia, and EPC levels in periph-

2 JOURNAL OF VASCULAR SURGERY Volume 50, Number 3 Yu et al 609 eral blood are low in a number of conditions in which poor circulation and vascular ischemia are present. 5 Obtaining enough EPCs to inject into ischemic tissue for treatment poses a problem: obtaining them from bone marrow requires anesthesia, and obtaining enough to inject from peripheral blood requires 5 to 6 liters of blood. 6,7 Cell culture has therefore been one way of expanding the supply of EPCs. SDF-1 and VEGF act on different mechanisms involved in neovascularization. It is possible that using them together may have additive or even synergistic effects. The study reported here used local SDF-1 injection into the ischemic hind legs of nude mice to attract intravenously injected EPCs that had been transduced with VEGF as a way to stimulate neovascularization of the ischemic tissue. The object was to see if the use of SDF and VEGF together was synergistic in producing neovascularization and in increasing blood flow to that area. The study also investigated whether a synergistic effect between these two substances occurred, thus increasing the migration of EPCs or decreasing the apoptosis of these cells that is seen when they are serum-starved. MATERIALS AND METHODS Human EPC isolation and culture. Ex vivo expansion of EPCs was performed as described by others. 5 Briefly, total human peripheral blood mononuclear cells were isolated from healthy human volunteers by densitygradient centrifugation with Histopaque-1077 (Sigma) and plated on culture dishes coated with human fibronectin (Sigma). The cells were cultured in endothelial cell basal medium-2 (EBM-2, Clonetics) supplemented with 20% fetal bovine serum (FBS), human VEGF-A, human fibroblast growth factor-2, human epidermal growth factor, insulin-like growth factor-1, ascorbic acid, and antibiotics. After 4 days in culture, nonadherent cells were removed by washing with phosphate-buffered saline (PBS), new medium was applied, and the culture was maintained through day 7. Examination of cells from the 7-day culture showed an acetylated low-density lipoprotein/lectin from Ulex europaeus agglutinin double-positive rate higher than 90%, and virtually 100% of cells were positive for von Willebrand factor. Flow cytometry analysis of the 7-day culture showed the percentage of CD133-positive cells was 23.2% 6.9%. These results suggested that almost all of the cells in the 7-day culture were of endothelial lineage, and, as such, they constituted an ex vivo expanded EPC-enriched fraction. EPC gene transfer. To obtain cells for overexpression of VEGF experiments after 7 days in culture, EPCs were transduced with pbabe containing human VEGF 165 complimentary DNA (Genebank accession no. AF486837; Td/V-EPCs) or with pbabe wild-type (Td/p-EPCs) retroviral vectors. Conditions and procedures for this transduction had previously been optimized in vitro to achieve top transfection efficiency, maximum protein expression, and highest cell viability. To determine the efficiency of the gene transfer, VEGF-EPCs transduced at 10 pfu/cell were subjected to flow cytometry. Transfection efficiency reached 61.3% in these Td/V-EPCs. After transfection, VEGF and wild-type viral containing supernatants were collected, filtered with m filters, and immediately frozen. Infections of EPCs were performed by mixing in a 1:1 ratio viral supernatants and EBM-2, supplemented with 8- g/ml polybrene (Sigma). After 3 hours of transduction, cells were washed with PBS and incubated with EPC media for 24 hours before transplantation. Migration assay. To investigate EPC migration activity, a modified Boyden chamber assay was performed using a 48-well microchemotaxis chamber (NeuroProbe). In brief, 30 L of SDF-1 (R & D Systems) diluted to 100 ng/ml or 30 L of EBM-2 alone supplemented with 0.1% bovine serum albumin (BSA), was placed in the lower compartment of a Boyden chamber. Cultured EPCs, Td/ V-EPCs, and Td/p-EPCs were harvested, cells were suspended in 50 L of EBM-2 supplemented with 0.1% BSA and antibiotics, then reseeded in the upper compartment. In experiments in which the SDF-1 receptor, CXCR4, was disabled, Td/V-EPCs were incubated for 30 minutes at 37 C with 5- g/ml of the CXCR4-specific antagonist AMD3100 (Sigma) before being reseeded. After incubation for 24 hours at 37 C, the Boyden chamber filter was removed, and the cells on the filter were counted manually in random high-power fields ( 100) in each well. All groups were studied at least in triplicate. Measurement of apoptosis by annexin V analysis. EPC apoptosis induced by serum starvation was quantified to determine whether SDF-1 and VEGF exerted a synergistic effect on EPC survival. In this protocol, cultured EPCs, Td/V-EPCs, and Td/p-EPCs were reseeded onto 6-well plates. After incubation for 24 hours, culture medium was removed and replaced with 1 ml of 3% serum EBM-2 supplemented with 100 ng/ml of SDF-1 or with 3% serum medium alone. The culture was maintained for 48 hours in serum deprivation. Annexin V assays were then performed according to the manufacturer s protocol (Annexin V-FITC Apoptosis Detection Kit, EMD Biosciences). The percentage of apoptotic cells was determined using a FACS Calibur flow cytometer equipped with Cell Quest software (BD Immunocytometry Systems). EPCs transplantation animal model. The combined effect of local administration of SDF-1 and VEGFtransduced EPC transplantation on therapeutic neovascularization was investigated in a murine model of hind limb ischemia. All procedures were performed in accordance with the Institutional Animal Care and Use Committee of Sun Yat-Sen University. Male nude mice (BALB/C-nu/ nu, Animal Centre of Sun Yat-Sen University), aged 8 to 10 weeks and weighing 18 to 22 g, were anesthetized with intraperitoneal sodium pentobarbital (160 mg/kg) for operative resection of one femoral artery and subsequently for laser Doppler perfusion imaging. A flow chart for the animal experiments is shown in Fig 1. The study comprised 60 mice divided into six groups of 10 mice in each group to receive various treatments. On day 1, after excision of the femoral artery in one hind limb,

3 610 Yu et al JOURNAL OF VASCULAR SURGERY September 2009 Fig 1. Flow chart of animal protocol. EBM-2, endothelial cell basal medium-2; EPC, endothelial progenitor cells; PBS, phosphate-buffered saline; SDF, stromal-derived factor. nude mice, in which angiogenesis is characteristically impaired, received an intramuscular injection of 1 g SDF-1 or of PBS in the center of the lower calf muscle of that hind limb. This was followed immediately by an intravenous injection into the tail vein of one of the following: culture-expanded Td/V-EPCs, Td/p-EPCs in 100 L EBM-2 media, or 100 L EBM-2 media without growth factors. To track the fate of the transplanted EPCs, 20 mice (5 mice in each EPC cohort) were injected with EPCs that had been labeled with the fluorescent carbocyanine DiI dye (Molecular Probes). The control group was 10 mice (5 mice in each EBM-2 cohort) that were injected with EBM-2. Before transplantation, EPCs in suspension were washed with PBS and incubated with DiI at a concentration of 2.5 g/ml PBS for 5 minutes at 37 C and 15 minutes at 4 C. After being washed twice in PBS, the cells were resuspended in EBM-2. At 30 minutes before euthanization on days 3 and 7, 5 mice in each group received an intravenous injection of 50 g ofbandeiraea simplicifolia lectin-1 (BS-1 lectin, Vector Laboratories) to identify the mouse vasculature. Plasma VEGF levels. To confirm that the Td/VEGF could mediate successful gene transfer at the protein level, an enzyme-linked immunosorbent assay (VEGF ELISA kit, Merck) was used to quantify VEGF levels in plasma from animals 1, 3, 7, and 28 days after intravenous injections of Td/V-EPCs or Td/p-EPCs. The results were compared with a standard curve constructed with VEGF, with each assay done in duplicate for each animal. Absorbance was measured at 450 nm by means of a microplate reader. RNA extraction and reverse transcription-polymerase chain reaction analysis. RNA was extracted from cultured EPCs, Td/V-EPCs, and Td/p-EPCs using TRIzol reagent (Invitrogen) according to the manufacturer s instructions. Reverse transcription-polymerase chain reaction (RT- PCR) of the CXCR4 gene and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene, as an internal control, was performed using 1 g of total RNA. PCR was performed for 35 cycles for CXCR4 and 25 cycles for GAPDH, with each cycle consisting of 94 C for 30 seconds and 64 C for 3 minutes. Amplification was done in 20- L reaction mixtures containing 0.4 U Taq polymerase. The following oligonucleotides were used as primers: for CXCR4 forward, 5=-TTCTACCCCAATGACTTGTG-3= and reverse, 5=-ATGTAGTAAGGCAGCCAACA-3=; for GAPDH forward, 5=-CACCATGGAGAAGGCTGG-3= and reverse, 5=-CTCAGTGTAGCCCAGGATGC-3=. Physiologic assessment of animals. Laser Doppler perfusion imaging (Perimed) was used to record blood flow measurements on days 1, 3, 5, 7, 14, 21, and 28 after surgery. In these digital color-coded images, a red hue indicates regions with maximum perfusion (the postimaging process blanched the red color somewhat, so in the photographs it is not as sensitive a detector as the other colors), a yellow hue indicates those with medium perfusion, and a blue hue those with lowest perfusion. The resulting images display absolute values in readable units. For quantification, the ratio of readable units in ischemic to nonischemic hind limb is determined. Immunofluorescent and histologic assessment. Tissue sections from the lower calf muscles of ischemic and healthy limbs were harvested on days 3, 7, and 28. To examine Td/V-EPC incorporation early after transplantation (at days 3 and 7) and SDF-1 effect on host endothelial cells, tissues from the mice injected with DiI-labeled EPCs and BS-1 lectin were embedded for frozen section samples. A total of 20 different fields (4 cross-sections from

4 JOURNAL OF VASCULAR SURGERY Volume 50, Number 3 Yu et al 611 Table. VEGF plasma levels after administration of Td/V-EPCs and Td/p-EPCs EPC, pg/ml Day 1 Day 3 Day 7 Day 28 Td/V-EPCs Td/p-EPCs EPC, Endothelial progenitor cells; VEGF, vascular endothelial growth factor. each animal) were randomly selected, and the DiI-labeled EPCs were counted ( 40 magnification). Capillary density in light microscopic sections was measured to assess the extent of neovascularization at day 28. Paraffin-embedded sections of 5- m thickness were stained for the mouse endothelial cell marker isolectin B4 (Vector Laboratories) and counterstained with eosin to detect capillary endothelial cells, as previously described. 8 Capillaries were counted ( 40 magnification) in 20 different fields that were randomly selected (2 or 3 cross-sections from each animal). Statistical analysis. All results are expressed as mean standard error of the mean. Statistical significance was evaluated using the unpaired t test for comparisons between two means. Multiple comparisons between three groups were done by analysis of variance. A value of P.05 denoted statistical significance. RESULTS Murine model of hind limb ischemia Transgene expression after transplantation of genemodified EPCs. The success of Td/VEGF-mediated gene transfer and expression was confirmed by using an ELISA assay to measuring VEGF levels in plasma samples obtained from nude mice after Td/V-EPC transplantation. These mice had significantly higher plasma VEGF levels on days 1, 3, 7, and 28 (all P.001) than mice that received EPCs containing the viral vector alone (Table). EPC incorporation into ischemic hind limb neovasculature. To study the effect of SDF-1 and VEGF on recruitment of EPCs from the systemic circulation, we measured the incorporation of injected Td/V-EPCs into the microvasculature of the ischemic limb and also the number of host endothelial cells found there. Transplanted human EPCs labeled with DiI were identified in tissue sections by red fluorescence, whereas the native mouse vasculature stained by premortem BS-1 lectin administration was identified by green fluorescence in the same tissue sections (Fig 2, A). Locally injected SDF-1 in combination with systemically injected VEGF-modified EPCs (the Td/V-EPC SDF-1 group) produced significantly increased DiI-labeled, red fluorescent EPC-derived vasculature and increased native mouse vasculature compared with the non-vegf-containing (Td/p-EPCs or EPCs) groups or the VEGF-containing (Td/V-EPC) group not supplemented by SDF-1 (Fig 2). When transplanted EPC-derived vasculature and mouse vasculature were both analyzed quantitatively in the same microscopic field, the density of DiI-labeled EPCs in these hind limb muscle tissue sections was greater in the VEGF EPC plus SDF-1 group, than in VEGF-lacking EPCS either with or without SDF-1, and in VEGFmodified EPCs alone (Td/V-EPCs plus SDF-1 group, cells/mm 2 ; Td/p-EPCs plus SDF-1, cells/mm 2 ; Td/p-EPCs, cells/mm 2 ; Td/V- EPCs, cells/mm 2, P.001; Fig 2, B). The density vasculature positive for BS-1 lectin in these tissue samples was also greater in the Td/V-EPCs plus SDF-1 group ( cells/mm 2 ) than in the other five groups examined (control, cells/mm 2 ; SDF-1, cells/mm 2 ; Td/p-EPCs, cells/mm 2 ; Td/p- EPCs plus SDF-1, cells/mm 2 ; and Td/V- EPCs, cells/mm 2, P.001; Fig 2, C). Physiologic and histologic assessment of animals given transplantation. After systemic human EPC transplantation with local intramuscular administration of either SDF-1 or saline, serial measurements of hind limb perfusion by Laser Doppler perfusion imaging were performed at days 1, 3, 5, 7, 14, 21, and 28. These measurements disclosed profound differences in the limb perfusion seen 28 days after induction of limb ischemia (Fig 3, A). The ratio of ischemic/normal blood flow in the Td/V-EPCs plus SDF-1 group indicated significantly greater hind limb perfusion than that seen in any other group, either mice transplanted with Td/p-EPCs or Td/V-EPCs alone, mice transplanted with Td/p-EPCs and given local injections of SDF-1, or mice given local administration of SDF-1 and no EPCs (10 for each group; P.01; Fig 3, B). Thus, the homing effect of local SDF-1 injection, as documented by counting EPC numbers in ischemic tissue, was accompanied by physiologic evidence for enhanced neovascularization, suggesting that the Td/V-EPCs that were attracted to the ischemic limb by SDF-1 and that increased the local VEGF levels there were subsequently incorporated into the developing vasculature. Histologic examination for capillary density was performed to provide anatomic evidence of vasculature increased by Td/V-EPCs in the SDF-1 -treated limbs. To quantify capillary density, staining with the endothelial cell marker isolectin B4 was performed on skeletal muscle sections retrieved from the ischemic hind limbs of mice on day 28. Capillary density, an index of neovascularization, was significantly higher in the Td/V-EPCs plus SDF-1 group ( capillaries/mm 2 ) than in the other groups (no treatment, capillaries/mm 2 ; SDF-1, capillaries/mm 2 ; Td/p-EPCs, capillaries/mm 2 ; Td/p-EPCs plus SDF-1, capillaries/mm 2 ; Td/V-EPCs, capillaries/mm 2 ; 10 in each group; P.01; Fig 3, C).

5 612 Yu et al JOURNAL OF VASCULAR SURGERY September 2009 Fig 2. Histologic identification of incorporation of endothelial progenitor cells (EPCs) into ischemic hind limb neovasculature. A, Representative photomicrographs show double fluorescence in ischemic limb muscles at day 7. Transplanted human EPCs labeled with DiI were identified in tissue sections by red fluorescence, and host mouse vasculature was identified by green fluorescence. Transplanted EPC-derived vasculature and mouse vasculature were analyzed quantitatively in the same microscopic field. B, Quantitative analysis of incorporated EPCs showed the density of DiI-labeled EPCs in tissue sections of hind limb muscles was greater in the Td/V-EPCs plus stromal-derived factor-1 (SDF-1 ) group than any of the other five groups (*P.001). C, Quantitative analysis of host endothelial cells. The density of Bandeiraea simplicifolia lectin-1 (BS-1) lectin positive vasculature in tissue sections of hind limb muscles was greater in the Td/V-EPCs plus SDF-1 group than in the other five groups (*P.001).

6 JOURNAL OF VASCULAR SURGERY Volume 50, Number 3 Yu et al 613 Fig 3. Physiologic and histologic evidence for enhanced neovascularization in ischemic hind limb. A, Laser Doppler perfusion imaging (LDPI) disclosed profound differences in the limb perfusion 28 days after induction of limb ischemia. In these digital color-coded images, red hue indicates regions with maximum perfusion, yellow indicates medium perfusion values, and blue designates the lowest perfusion. B, In quantitative analysis of perfusion recovery measured by LDPI, the ratio of ischemic/normal blood flow in the group that received Td/V-endothelial progenitor cells (EPCs) plus stromal-derived factor-1 (SDF-1 ) was greater than in the other five groups (10 in each group; *P.01). C, Quantitative analysis of capillary density in light microscopic sections at day 28 showed capillary density of the Td/V-EPCs plus SDF-1 group was significantly higher than other groups (10 in each group; *P.01).

7 614 Yu et al JOURNAL OF VASCULAR SURGERY September 2009 Fig 4. Stromal-derived factor-1 (SDF-1 ) and vascular endothelial growth factor (VEGF) synergize to promote angiogenic function of endothelial progenitor cells (EPCs) in vitro. A and B, SDF-1 induced a notable migration of EPC that was significantly more efficient in the presence of VEGF expression in Td/V-EPCs groups compared with Td/V-EPCs or SDF-1 alone. SDF-1 -mediated cell migration in Td/V-EPCs group was completely inhibited by AMD3100, the CXCR4-specific antagonist (**P.01, ***P.001). EPCs, Td/V-EPCs or Td/p-EPCs were seeded for transwell migration assay in the presence of SDF-1 respectively. After 24 hours of incubation, migration rates were quantified by counting the migrated cells in five random fields (original magnification 100.) Data summarize three independent experiments. One representative of three independent experiments was shown. The range bars show the standard error of the mean. C, Reverse transcription-polymerase chain reaction analysis of messenger RNA transcripts for CXCR4 in EPCs, Td/V-EPCs, or Td/p-EPCs respectively. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal control. D, SDF-1 and VEGF synergistically protect EPC apoptosis in the Td/V-EPCs plus SDF-1 group. EPCs, Td/V-EPCs, or Td/p-EPCs were cultured with medium containing 3% and 20% fetal bovine serum (FBS), respectively, for 24 hoursy. SDF-1 (100 ng/ml) was added into the culture. Cellular apoptosis was analyzed by annexin V and propidium iodide staining and expressed as the percentage of apoptotic cells. Range bars designate the standard error of the mean. In vitro results SDF-1 and VEGF synergize to promote angiogenic function of EPCs. EPC migration is a critical step in angiogenesis. 9 We studied the directional migration of EPCs induced by SDF-1 and VEGF. SDF-1 significantly increased migration compared with medium in all three EPC groups (P.001, Fig. 4, A and B). The combination of SDF-1 and Td/V-EPC resulted in significantly more

8 JOURNAL OF VASCULAR SURGERY Volume 50, Number 3 Yu et al 615 migration than that seen with Td/V-EPC or SDF-1 alone ( /mm 2, /mm 2, and /mm 2, respectively; P.001; Fig 4, A and B). We next examined the synergistic mechanism by which SDF-1 and VEGF increased migration. Preincubation with the CXCR4-specific antagonist AMD3100 completely blocked the cell migration seen with the SDF-1 and Td/V-EPC combination, confirming that the increased migration seen with this combination was mediated through the SDF-1 receptor CXCR4 (P.001; Fig 4, B). Further experiments showed that CXCR4 mrna levels were increased in Td/V-EPCs (P.001 compared with EPCs or Td/p-EPCs; Fig 4, C), indicating that VEGF increased the SDF-1 mediated migration of EPCs through up-regulating CXCR4. VEGF and SDF-1 synergistically protected EPC from apoptosis. We hypothesized that SDF-1 and VEGF synergistically protected EPCs from apoptosis. To test this hypothesis, fresh EPCs were cultured in medium containing low or high concentrations of FBS. Serum starvation (3% FBS) caused apoptosis in 50% of EPCs, a significantly greater amount (P.01) than the 10% apoptosis seen in 20% serum. SDF-1 and Td/V-EPCs, when used separately, independently and marginally decreased the percentage of apoptotic cells induced by sera starvation (P.01, compared with EPCs or Td/p-EPCs alone; Fig 4, D). Strikingly, however, Td/V-EPCs together with SDF-1 reduced the percentage of apoptotic cells induced by 3% FBS almost to the levels seen with no serum starvation (P.001, compared with Td/V-EPCs or SDF-1 alone; Fig 4, D), suggesting that VEGF and SDF-1 act synergistically in protecting EPCs from apoptosis. These in vitro data, taken together, indicate that multiple mechanisms are implicated in the in vivo synergistic angiogenic action of SDF-1 and VEGF. DISCUSSION The results of this study show that combining local administration of SDF-1 with intravenous delivery of EPCs that over-express hvegf 165 is superior to the use of either substance alone in revascularizing ischemic tissues. In our hands, combination therapy significantly increases EPC mobilization, endothelial cell proliferation, and blood flow, and decreases apoptosis compared with the use of either agent alone. SDF-1 and VEGF have different major actions in increasing angiogenesis. SDF-1 acts as a chemoattractant for EPCs and suppresses their apoptosis. 6,10 Its mitogenic activity is limited. 4 VEGF is a mitogen that increases cell proliferation. 11 By itself, it has only a weak chemoattractant activity. Local injection of SDF-1 will increase the number of EPCs drawn to the ischemic site. VEGF over-expression by the EPCs increases the number of SDF-1 receptors (ie, CXCR4 molecules) 12 on the EPCs and in this way further increases their response to SDF-1. SDF-1 increases the VEGF-induced proliferative response of EPCs by decreasing apoptosis in these cells. This cytokine also induces VEGF release from the EPCs, 12 thus further increasing the mitogenic response to VEGF. Theoretic reasons to use EPCs that have been enriched in VEGF through gene transfer are that this will increase the ability of the local SDF-1 injection to attract EPCs, and VEGF, being produced locally by VEGF-enriched EPCs instead of being injected systemically, will be present in higher local concentrations and for a longer time, and this will also increase its mitogenic activity. 13 Local gene therapy has been thought to be better than intravenous protein therapy, because the protein will be expressed longer and because the therapy can be delivered locally instead of systemically. 14 But clinical trials using plasmids containing VEGF have produced mixed results. 14,15 Gene transfer of SDF-1 through plasmids injected into the ischemic hind legs of mice, however, increased the ischemic/normal blood flow ratio from approximately 0.35 (plasmid alone) to 0.70 (plasmid with SDF-1 ). 16 Our results using local injection of SDF-1 instead of the SDF-1 gene-containing plasmid showed no effect on the blood flow ratio of SDF-1 alone (ratio about 0.20 for both SDF-1 and saline injection), although the combination of SDF-1 and VEGF gene-modified EPCs resulted in a blood flow ratio of about 0.70, comparable to that seen in the gene transfer experiment with SDF-1 plasmids. Iwaguro et al, 17 in another study with a protocol similar to ours, injected EPCs with VEGF-containing plasmids into the tail veins of nude mice with ischemic hind limbs and measured the ischemic/normal blood flow ratio at day Their ratio of approximately 0.70 was similar to the 0.70 ratio seen at 28 days in our doubly treated mice and greater than the 0.40 ratio in our mice treated only with the VEGF plasmid containing EPCs. SDF-1 is said to act in both chemotaxis and apoptosis by binding to the CXCR4 receptor, thus activating a tyrosine kinase coupled to an inhibitory G protein. 9 VEGF stimulation increases nitric oxide. 10 Our research did not address the specific intracellular mechanisms involved in the synergism seen between SDF-1 and VEGF. But it did show that this synergism was seen in all actions examined increase in blood flow, increase in cell migration, and decrease in apoptosis. The unanswered question is whether the same cellular mechanisms drive the synergy in each action. Another question is how the progenitor cells described by Wragg et al, 18 which do not enter the endothelium and proliferate but are thought to play a paracrine role in angiogenesis, fit into previously described studies of angiogenesis. This study has several limitations. Controls such as VEGF and a nonchemotactic cytokine such as IP-10 were not included in the migration assay. We did not determine whether the injected EPCs were taken-up by organs such as the spleen or liver, nor did we determine whether blocking the CXCR4 receptor inhibited the SDF-1 effect on cell migration or whether the VEGF transfection merely had a nonspecific effect on CXCR4 expression. We also did not have a detectable marker on the retroviral marker and so could not determine why VEGF levels declined rapidly

9 616 Yu et al JOURNAL OF VASCULAR SURGERY September 2009 after transduction. We hope to investigate some of these questions in future studies. CONCLUSIONS We have shown for the first time, to our knowledge, that combining locally injected SDF-1 with intravenous delivery of EPCs over-expressing hvegf 165 is superior to either of these strategies used alone in increasing the blood supply to ischemic tissues. Our results suggest that the synergistic SDF-1 plus VEGF angiogenic pathway is critical for the success of EPC-induced neovascularization and that simultaneously targeting both SDF-1 and VEGF signals may be a novel and efficient strategy for treating ischemic diseases. We thank Dr Gui-Fu Wu and D. Qiang Xie at the Division of Cardiology, The First Affiliated Hospital of Sun Yat-Sen University for their help in animal experiments. AUTHOR CONTRIBUTIONS Conception and design: JY, XH, WL, SW Analysis and interpretation: JY, XH Data collection: JY, XH, XP, HZ Writing the article: JY, XH Critical revision of the article: JY, XH, WL, CY, LX Final approval of the article: JY, XH, SW Statistical analysis: JY, XH Obtained funding: SW Overall responsibility: SW JY and XH contributed equally to this work. REFERENCES 1. De Falco E, Porcelli D, Torella AR, Straino S, Iachininoto MG, Orlandi A, et al. SDF-1 involvement in endothelial phenotype and ischemiainduced recruitment of bone marrow progenitor cells. Blood 2004;104: Heeschen C, Lehmann R, Honold J, Assmus B, Aicher A, Walter DH, et al. Profoundly reduced neovascularization capacity of bone marrow mononuclear cells derived from patients with chronic ischemic heart disease. Circulation 2004;109: van Weel V, Seghers L, de Vries MR, Kuiper EJ, Schlingemann RO, Bajema IM, et al. Expression of vascular endothelial growth factor, stromal cell-derived factor-1, and CXCR4 in human limb muscle with acute and chronic ischemia. Arterioscler Thromb Vasc Biol 2007;27: Salcedo R, Wasserman K, Young HA, Grimm MC, Howard OM, Anver MR, et al. Vascular endothelial growth factor and basic fibroblast growth factor induce expression of CXCR4 on human endothelial cells: in vivo neovascularization induced by stromal-derived factor-1alpha. Am J Pathol 1999;154: Kalka C, Masuda H, Takahashi T, Kalka-Moll WM, Silver M, Kearney M, et al. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization. Proc Natl Acad Sci U S A 2000;97: Yamaguchi J, Kusano KF, Masuo O, Kawamoto A, Silver M, Murasawa S, et al. Stromal cell-derived factor-1 effects on ex vivo expanded endothelial progenitor cell recruitment for ischemic neovascularization. Circulation 2003;107: Jin DK, Shido K, Kopp HG, Petit I, Shmelkov SV, Young LM, et al. Cytokine-mediated deployment of SDF-1 induces revascularization through recruitment of CXCR4 hemangiocytes. Nat Med 2006;12: Kawamoto A, Gwon HC, Iwaguro H, Yamaguchi JI, Uchida S, Masuda H, et al. Therapeutic potential of ex vivo expanded endothelial progenitor cells for myocardial ischemia. Circulation 2001;103: Oh IY, Yoon CH, Hur J, Kim JH, Kim TY, Lee CS, et al. Involvement of E-selectin in recruitment of endothelial progenitor cells and angiogenesis in ischemic muscle. Blood 2007;110: Broxmeyer HE, Kohli L, Kim CH, Lee Y, Mantel C, Cooper S, et al. Stromal cell-derived factor-1/cxcl12 directly enhances survival/antiapoptosis of myeloid progenitor cells through CXCR4 and G(alpha)i proteins and enhances engraftment of competitive, repopulating stem cells. J Leukoc Biol 2003;73: Shi Q, Rafii S, Wu MH, Wijelath ES, Yu C, Ishida A, et al. Evidence for circulating bone marrow-derived endothelial cells. Blood 1998;92: Ruiz de Almodovar C, Luttun A, Carmeliet P. An SDF-1 trap for myeloid cells stimulates angiogenesis. Cell 2006;124: Kryczek I, Lange A, Mottram P, Alvarez X, Cheng P, Hogan M, et al. CXCL 12 and vascular endothelial growth factor synergistically induce neoangiogenesis in human ovarian cancers. Cancer Res 2005;65: Rajagopalan S, Mohler ER 3rd, Lederman RJ, Mendelsohn FO, Saucedo JF, Goldman CK, et al. Regional angiogenesis with vascular endothelial growth factor in peripheral arterial disease: a phase II randomized, double-blind, controlled study of adenoviral delivery of vascular endothelial growth factor 121 in patients with disabling intermittent claudication. Circulation 2003;108: Kusumanto YH, van Weel V, Mulder NH, Smit AJ, van den Dungen JJ, Hooymans JM, et al. Treatment with intramuscular vascular endothelial growth factor gene compared with placebo for patients with diabetes mellitus and critical limb ischemia: a double-blind randomized trial. Hum Gene Ther 2006;17: Hiasa K, Ishibashi M, Ohtani K, Inoue S, Zhao Q, Kitamoto S, et al. Gene transfer of stromal cell-derived factor-1alpha enhances ischemic vasculogenesis and angiogenesis via vascular endothelial growth factor/ endothelial nitric oxide synthase-related pathway: next-generation chemokine therapy for therapeutic neovascularization. Circulation 2004; 109: Iwaguro H, Yamaguchi J, Kalka C, Murasawa S, Masuda H, Hayashi S, et al. Endothelial progenitor cell vascular endothelial growth factor gene transfer for vascular regeneration. Circulation 2002;105: Wragg A, Mellad JA, Beltran LE, Konoplyannikov M, San H, Boozer S, et al. VEGRF1/CXCR4-positive progenitor cells modulate local inflammation and augment tissue perfusion by a SDF-1-dependent mechanism. J Mol Med 2008;86: Submitted Nov 25, 2008; accepted May 13, 2009.