Isolation and Transplantation of Autologous Circulating Endothelial Cells Into Denuded Vessels and Prosthetic Grafts
Implications for Cell-Based Vascular Therapy
Background— Blood-borne endothelial cells originating from adult bone marrow were reported previously. These cells have the properties of an endothelial progenitor cell (EPC) and can be mobilized by cytokines and recruited to sites of neovascularization, where they differentiate into mature endothelial cells. Current protocols for isolation of EPCs from peripheral blood rely on enrichment and selection of CD34+ mononuclear cells.
Methods and Results— In this report, we describe a streamlined method for the isolation and expansion of EPCs from peripheral blood and evaluate their therapeutic potential for autologous cell-based therapy of injured blood vessels and prosthetic grafts. A subset of unfractionated mononuclear cells exhibited the potential to differentiate in vitro into endothelial cells under selective growth conditions. The cells were efficiently transduced ex vivo by a retroviral vector expressing the LacZ reporter gene and could be expanded to yield sufficient numbers for therapeutic applications. Transplantation of these cells into balloon-injured carotid arteries and into bioprosthetic grafts in rabbits led to rapid endothelialization of the denuded vessels and graft segments, resulting in significant reduction in neointima deposition.
Conclusions— We conclude that transplantation of EPCs may play a crucial role in reestablishing endothelial integrity in injured vessels, thereby inhibiting neointimal hyperplasia. These findings may have implications for novel and practical cell-based therapies for vascular disease.
Received October 23, 2001; de novo received May 16, 2003; revision received July 24, 2003; accepted July 28, 2003.
Several reports have documented the existence of blood-borne endothelial cells originating in adult bone marrow.1–4 These endothelial-lineage cells have the properties of an endothelial progenitor cell (EPC) and are recruited to foci of neovascularization, such as in ischemic muscle5 and myocardium,6 where they differentiate into mature endothelial cells, thereby suggesting a role in postembryonic vasculogenesis. The relative abundance of EPCs is low in basal conditions, but the number of circulating cells is increased severalfold after exogenous mobilization with cytokines.7
The potential of these cells as therapeutic vehicles for tissue salvage from ischemic damage has been suggested.8,9 Recently, statin therapy was found to accelerate reendothelialization and to attenuate neointima formation in denuded carotid arteries because of enhanced EPC mobilization from the bone marrow,10 suggesting that EPC recruitment to sites of injury may participate in tissue repair. However, the suitability of these cells in vascular therapy has not been fully evaluated. Kaushal et al11 reported that transplantation of EPCs onto decellularized porcine iliac artery leads to reconstitution of a bioactive endothelial layer and sustained patency when these preparations are transplanted as carotid interpositional grafts. These observations have implications for the design of cell-based therapies for vascular diseases. Revascularization procedures such as percutaneous balloon angioplasty and bypass grafting are widely used in the treatment of coronary artery disease but are often prone to failure because of restenosis, thrombosis, and vasospasm.12 Endothelial cell loss is a major contributing factor to postangioplasty restenosis and graft failure.13 In this regard, autologous EPC transplantation may be a feasible adjunctive therapeutic option for accelerated reendothelialization of vessels and grafts injured during revascularization procedures and prevention of neointima hyperplasia.
In the present study, we describe a streamlined method for rapid isolation, growth, and ex vivo expansion of EPCs from peripheral blood of rabbits and evaluate the ability of these cells to reendothelialize and inhibit neointimal hyperplasia in injured carotid arteries and prosthetic grafts. Our results indicate that EPC transplantation leads to rapid reendothelialization of denuded arteries and prosthetic grafts, resulting in significant inhibition of neointima thickening. Furthermore, these cells are amenable to transduction with high-titer retroviral vectors and may be useful for genetic engineering of cell-based therapies for vascular diseases.
Adult New Zealand White male rabbits (body weight, 3 to 3.5 kg) were purchased from Millbrook Breeding Laboratory (Amherst, Mass) and maintained on a 12:12 hour light:dark cycle in an ambient temperature of 24°C and 60% humidity. Food and water were provided ad libitum. The Harvard Medical Area Standing Committee on Animal Care approved all animal procedures.
Animals were anesthetized with a mixture of ketamine (25 mg/kg) and xylazine (10 mg/kg) administered intravenously and supplemented as needed. A left groin incision was made under sterile conditions. The femoral vein was catheterized, and 50 mL of blood was withdrawn into a heparinized syringe. The volume of blood was instantly replaced by an equivalent volume of normal saline. The wound was closed, and a protective collar was placed for 5 days.
Cell Isolation and Culture Conditions
Blood was diluted with an equal volume of HBSS, and mononuclear cells were separated by density-gradient centrifugation using 1.077 g/mL Histopaque solution (Sigma Chemicals). The whole buffy coat was plated on 100-mm dishes coated with recombinant fibronectin (PanVera). Cells were maintained on endothelial growth media supplemented with EGM-2MV Single Quots (Clonetics). First-passage cells were used for all in vivo applications.
For immunohistochemistry, first-passage cells were plated on 4-well chamber slides and fixed with 4% paraformaldehyde for 30 minutes. Slides were blocked with 3% BSA in PBS-T for 1 hour and incubated overnight at 4°C in 1:100 dilution of goat polyclonal anti–von Willebrand factor (DiaSorin), Flk-1, or VE-cadherin (Santa Cruz Biotechnology). Horseradish peroxidase activity was visualized with DAB. All other antigens were detected by incubation in 1:500 dilution of Alexa Fluor 546 rabbit anti-goat IgG conjugate (Molecular Probes) for 1 hour at room temperature. To test binding and uptake of LDL particles, cells were incubated with 10 μg/mL Dil-Ac-LDL (Biomedical Technologies) for 4 hours at 37°C. Vascular tube assay was performed by plating 40 000 cells in 1 well of a 24-well plate precoated with 300 μL matrigel (Becton Dickinson).
Retroviral Production and Transduction
The retroviral vector was constructed by inserting the bacterial β-galactosidase gene (LacZ) at the NcoI and BamHI sites of linearized pC.MMP plasmid. The pC.MMP vector construct is a derivative of pMFG(MPSV). The coding sequence is preceded by the SV40 large T-antigen nuclear localization signal. Generation and titering of VSV-G pseudotyped retroviral particles was carried out by the Harvard Gene Therapy Initiative. The titers were 0.5×109 to 1×109 IU/mL. First-passage EPCs at 30% to 40% confluence in 100-mm dishes were exposed to 100 μL of virus solution in 5 mL of complete medium for 6 hours in the presence of 8 μg/mL Polybrene (Aldrich Chemical). Cells were transplanted 4 days after transduction.
Balloon Injury Model
For arterial denudation, the animals were heparinized systemically (100 U/kg), and the right common carotid artery was exposed to the level of the bifurcation through a midline neck incision as previously described.14 A 2F Fogarty balloon catheter (Baxter) was inserted through the external carotid, inflated, and passed 3 times along the length of the isolated segment (3.5 to 4.5 cm in length). The segment was rinsed with saline and cannulated. Cells mixed in saline were instilled at ≈2500 cells/mm2 and incubated for 30 minutes. The animals were turned on their sides twice to promote uniform cell attachment. After the incubation, unbound cells were aspirated, and the catheter was removed. Control vessels were incubated with an equivalent volume of saline. The external carotid artery was tied off, and the neck was closed.
Tissue Collection and Preparation
Carotid segments were isolated at 4, 14, and 30 days after grafting. The segments were rinsed and cut into 8 to 10 fragments. Two fragments were fixed overnight in 10% formalin and embedded in paraffin by conventional methods. The remaining fragments were snap-frozen in OCT compound for preparation of fresh-frozen sections.
Cell Sodding and Implantation of Vascular Grafts
Gradient expanded polytetrafluoroethylene grafts, 4-mm ID with pore sizes of 60 μm on the luminal side and 20 μm on the outer side, were obtained from Atrium. Grafts were flushed with 50% ethanol in PBS, coated with fibronectin, and sodded with first-passage cells at 2500 cells/mm2 by passing the cell solution, in a volume of 3 mL, 3 times through the graft. Infiltrated grafts were maintained in organ culture under nonflow conditions for 48 hours before implantation as carotid interpositions.
For detection of LacZ expression, specimens were fixed in glutaraldehyde and stained with X-gal using a commercially available kit (InVitrogen). For en face detection of LacZ-positive cells, vessels were opened lengthwise and incubated in X-gal solution for 2 hours at 37°C. For histological analysis of endothelialization, 5-μm frozen sections were incubated in 1:200 dilution of monoclonal anti-CD31 antibody (Dako) overnight at 4°C, developed in DAB, and counterstained with hematoxylin. Endothelialization was calculated as the ratio of the surface covered by CD31-positive cells and the total luminal surface. Some representative sections were double-stained with anti-CD31 and X-gal to confirm the phenotype and origin of the grafted cells. The integrity of the endothelial lining was also visualized by scanning electron microscopy (SEM). For assessment of transplanted EPC proliferation in vivo, bromodeoxyuridine (BrdU) (35 mg/kg IP, Sigma) was injected into the rabbits at 24 and 12 hours before termination. Frozen sections were processed for BrdU staining with a commercial kit (Pharmingen). For morphometric analysis of neointima, 5-μm paraffin sections were stained with Accustain Elastic Stain (Sigma). Neointimal and medial cross-sectional areas were calculated in 6 to 8 individual sections by use of Adobe Photoshop.
Where applicable, results are presented as mean±SEM. Unpaired t test was used for comparisons between EPC-transplanted and saline-treated groups. A probability value ≤0.05 was considered to indicate statistical significance.
Isolation and Retroviral Transduction of Autologous Endothelial Cells From Peripheral Blood
Under the culture conditions used, we found colonies of adherent cells at approximately day 5 after plating. These islets proliferated over the course of the next few days, from a center of densely packed round cells to a periphery of seemingly mature and contact-inhibited cells showing the “cobblestone” appearance characteristic of an endothelial cell monolayer (Figure 1A). The number of colonies ranged between 5 and 30 per sample of blood, thus allowing us to estimate a frequency of 1 to 4 EPCs per 1×107 leukocytes. Repeated washing steps increased the purity of the culture by removing nonadherent cells and debris. Most cultures had reached confluence by 14 to 20 days, yielding 7600 to 54 000 cells/cm2. The cells could be passaged several times without significant alterations to their morphology or growth characteristics. After 1 week of culture, the majority of the cells expressed the endothelial markers von Willebrand factor (Figure 1B), Flk-1 (Figure 1C), and VE-cadherin (Figure 1D). The cells were also capable of taking up LDL particles from the media (Figure 1E) and assembled into primitive vascular tube-like structures when plated in Matrigel (Figure 1F). Exposure to retroviral particles expressing LacZ led consistently to transduction efficiencies >70%.
Repopulation of Freshly Denuded Carotid Artery
Figure 2 shows coverage of the luminal surface of unseeded and seeded balloon-injured carotid artery with EPCs expressing LacZ at 4, 14, and 30 days after grafting. Extensive coverage (>70%) of the luminal surface with LacZ-positive cells was seen 4 days after transplantation (Figure 2B). The LacZ-positive area was significantly reduced in vessels harvested 2 weeks after seeding (Figure 2D). No positive cells were detected in the denuded vessels 4 weeks after transplantation (Figure 2F) or at any time in unseeded vessels (Figure 2, A, C, and E). All seeded arteries appeared viable macroscopically.
Reendothelialization of Denuded Carotid Artery
Figure 3 shows endothelialization of denuded carotid artery after transplantation of EPCs). No evidence of LacZ-positive cells or of an endothelial monolayer was found in the unseeded vessel segments (Figure 3A). LacZ-positive cells were seen lining the lumen 4 days after transplantation (Figure 3B). Costaining with the endothelial cell marker CD31 confirmed the endothelial phenotype of the transplanted cells (Figure 3B). SEM of silver nitrate–stained vessels confirmed the absence of endothelium in the saline controls (Figure 3C) and revealed the integrity of the newly formed endothelial monolayer in the seeded vessels (Figure 3D) Total endothelial cell coverage was ≈60% in the seeded vessels and <5% in the unseeded vessels at day 4 after transplantation and was increased further at 4 weeks in both seeded and unseeded vessels(Figure 3E).
Proliferation of Transplanted EPCs in Balloon-Denuded Carotid Artery
We stained adjacent sections from both control and seeded vessels for BrdU 4 days after transplantation to determine whether the EPCs retain the ability to proliferate after grafting into the denuded vessels. BrdU-positive cells were detected in the adventitia and media but not in the intima of the unseeded vessels, which was devoid of endothelium (Figure 4, A and C). In contrast, clusters of BrdU-positive cells were seen in the intima of the seeded vessels, colocalizing with LacZ-positive cells (Figure 4, B and D), suggesting that some of the transplanted EPCs undergo proliferation in vivo. No BrdU-positive cells were seen at 2 and 4 weeks after transplantation (data not shown).
Inhibition of Neointimal Hyperplasia in Balloon-Injured Carotid Artery
To determine the ability of EPC transplantation to inhibit neointima deposition, seeded carotid arteries were isolated 2 weeks after balloon injury. No evidence of neointima was found in uninjured vessels (Figure 5A), whereas balloon injury led to the development of prominent neointima (Figure 5B). EPC transplantation reduced neointima deposition significantly (Figure 5C). Neointimal thickness and neointima/media ratio were significantly reduced in the seeded vessels compared with unseeded vessels (neointima thickness: EPC, 0.0353±0.0053 mm; saline, 0.0942±0.017 mm; neointima/media ratio: EPC, 0.29±0.051; saline, 0.57±0.034; P<0.05 for both).
Endothelialization of Vascular Bioprosthesis
We also evaluated the ability of EPCs to endothelialize vascular bioprostheses. Approximately 60% to 70% of the luminal surface of the sodded grafts was covered by endothelial cells before implantation (Table). Grafts harvested at 7, 14, and 28 days after implantation were patent. No positive cells were found in unsodded grafts (Figure 6A). X-gal staining showed patchy retention of the transplanted cells, covering between 5% and 15% of the surface in the sodded grafts (Figure 6C, Table). No evidence of endothelialization was found in the unsodded grafts by SEM (Figure 6C). The LacZ-positive area in the sodded grafts at 7 and 14 days was comparable to the degree of endothelialization seen by SEM (Figure 6D). Four weeks after sodding, 40% to 60% of the total area in the sodded grafts was covered by endothelial cells, compared with <5% in the unsodded grafts (Table).
The unique functional role of endothelial cells in maintenance of vessel homeostasis renders them optimal candidates for cell-based therapy for vascular disease. In this study, we describe a simple and cost-effective method for isolation, culture, and expansion of EPCs from the mononuclear fraction of peripheral blood that can be induced to differentiate into cells with endothelial cell–like properties. We consistently obtained a sufficient number of cells to allow expansion for in vivo applications within 2 to 3 weeks of initial plating. Furthermore, we show that these cells are amenable to transduction by retroviral vectors with high efficiency, thus underlying their potential use as vehicles for delivery of therapeutic genes at sites of injury.
Previously reported methods of harvesting and culturing EPCs have relied on magnetic bead or cytofluorometric selection for cells expressing CD34, vascular endothelial growth factor receptor, or AC133.15,16 Here, we show that endothelial-like cells can be selectively grown in culture from mononuclear cells on the basis of adherence and requirement for endothelial cell–specific growth conditions without any further enrichment steps. The present study does not identify the specific lineages of the hematopoietic precursor(s) of EPCs. Asahara et al1 showed that the EPCs originate predominantly from a CD34+ population. However, because we did not select for any specific markers, we cannot exclude the possibility that our heterogeneous mononuclear fraction contains more than 1 type of EPC precursor.
Our study demonstrates the successful implantation of autologous EPCs onto balloon-denuded rabbit arteries. We chose the rabbit model because it allows the collection of a large sample of blood required for ex vivo expansion of an adequate number of cells for subsequent autologous transplantation. Furthermore, the pathological time course of neointima formation after balloon injury is well characterized.17 Our results show that EPC transplantation leads to early and nearly complete reendothelialization of the denuded carotid artery, resulting in inhibition of neointimal proliferation. These finding may have implications for treatment of vascular proliferative disease, because restenosis after balloon angioplasty remains a major problem in vascular therapy.12 We have previously demonstrated the potential benefit of gene therapy strategies in stabilizing vessel wall function and limiting neointimal proliferation after injury.18,19 The ability to isolate, expand, and genetically modify EPCs from peripheral blood adds a new dimension for intervention and establishment of cell-based therapy for vein graft and angioplasty-related vascular injury. Rapid restoration of the endothelial lining and the ability to engineer these cells to generate a therapeutically enhanced surface might be a suitable alternative to direct gene therapies for restenosis. More significantly, peripheral blood is a nondepleting source of endothelial cells that could be used for autologous reendothelialization of injured vessels.
Our data indicate a time-dependent decrease in the area covered by the transplanted cells. Several mechanisms may account for the progressive loss in LacZ transgene expression. First, the decrease in LacZ-positive cells may be a result of rapid cell turnover with simultaneous replacement by native cells. The transplanted cells may initially provide a local environment for mobilization and homing of neighboring endothelial cells or native EPCs to the injured area, which in time may eventually replace the transplanted cells. Second, the loss of transgene expression could also be caused by silencing of transcription of the integrated proviral cassette.20 Interestingly, a similar time course of reendothelialization and transgene expression was reported by Conte et al,21 who showed that transplantation of retrovirally transduced autologous vein endothelial cells onto denuded arteries leads to rapid endothelialization of the vessels in the first week after transplantation and subsequent attenuation of reporter gene expression 2 weeks after transplantation. We postulate that the grafted cells play an essential role in promoting early reendothelialization of the injured vessels by engrafting and releasing chemokines and other mediators that may enhance mobilization and homing of endogenous cells for endothelialization.
Regarding the engineering of vascular bioprostheses, we introduce the concept of cell sodding as a strategy to reduce the time needed for endothelialization of prosthetic grafts. The use of gradient expanded polytetrafluoroethylene material with 60-μm pore size on the luminal side and 20-μm on the outer side seems to favor retention of the sodded cells. Our results are not as dramatic as those reported previously with autologous bone marrow cells,22 where near complete luminal coverage was reported. Nevertheless, we observed significant endothelialization (60% to 70% coverage) of the sodded grafts, despite the relatively low retention of the transplanted cells, suggesting that these cells may provide homing signals for recruitment of endogenous cells to the graft.
In conclusion, the present study describes a simplified method for the isolation, expansion, and genetic manipulation of EPCs from peripheral blood and demonstrates that autologous transplantation of these cells accelerates endothelialization of injured vessels and prosthetic grafts, leading to inhibition of neointima hyperplasia. Given the potential of genetic modification of these cells, these findings underline the therapeutic potential of EPCs as vectors for cell- and gene-based therapies for vascular disease.
This work was supported by National Institutes of Health grants HL-35610, HL-058516, HL-072010, and HL-073219. Dr Griese was supported by a postdoctoral research grant from the Deutsche Forschungsgemeinschaft. Dr Ehsan was the recipient of a postdoctoral fellowship from the American Heart Association. Dr Melo is a new investigator of the Heart and Stroke Foundation of Canada and is supported by grants from the Canadian Institutes of Health Research. Dr Dzau is the recipient of a National Heart, Lung, and Blood Institute MERIT Award.
↵*These authors contributed equally to this work.
Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science. 1997; 275: 964–967.
Shi Q, Rafii S, Wu MH, et al. Evidence for circulating bone marrow-derived endothelial cells. Blood. 1998; 92: 362–367.
Asahara T, Masuda H, Takahashi T, et al. Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in physiological and pathological neovascularization. Circ Res. 1999; 85: 221–228.
Kawamoto A, Gwon H-C, Iwaguro H, et al. Therapeutic potential of ex vivo expanded endothelial progenitor cells for myocardial ischemia. Circulation. 2001; 103: 634–637.
Shintani S, Murohara T, Ikeda I, et al. Augmentation of postnatal neovascularization with autologous bone marrow transplantation. Circulation. 2001; 103: 897–903.
Kalka C, Matsuda H, Takahashi T, et al. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization. Proc Natl Acad Sci U S A. 2000; 97: 3422–3427.
Walter DH, Rittig K, Bahlmann FH, et al. Statin therapy accelerates reendothelialization: a novel effect involving mobilization and incorporation of bone marrow-derived endothelial progenitor cells. Circulation. 2002; 105: 3017–3024.
Qian J-Y, Haruno A, Asada Y, et al. Local expression of C-type natriuretic peptide suppresses inflammation, eliminates shear stress–induced thrombosis, and prevents neointima formation through enhanced nitric oxide production in rabbit injured carotid arteries. Circ Res. 2002; 91: 1063–1069.
Peichev M, Naiyer AJ, Pereira D, et al. Expression of VEGFR-2 and AC133 by circulating human CD34+ cells identifies a population of functional endothelial precursors. Blood. 2000; 95: 952–958.
Yamaguchi TP, Dumont DJ, Conlon DJ. Flk-1 and flt-related tyrosine kinase is an early marker for endothelial cell precursors. Development. 1993; 118: 489–498.
von der Leyen HE, Gibbons GH, Morishita R, et al. Gene therapy inhibiting neointimal vascular lesion: in vivo transfer of endothelial cell nitric oxide synthase gene. Proc Natl Acad Sci U S A. 1995; 92: 1137–1141.
Chillita PM, Kohn DB. Lack of expression from a retroviral vector after transduction of murine hematopoietic stem cells is associated with methylation in vivo. Proc Natl Acad Sci U S A. 1994; 91: 2567–2571.
Conte MS, Birinyi LK, Miyata T, et al. Efficient repopulation of denuded rabbit arteries with autologous genetically modified endothelial cells. Circulation. 1994; 89: 2161–2169.
Bhattacharya V, McSweeney PA, Shi Q, et al. Enhanced endothelialization and microvessel formation in polyester grafts seeded with CD34+ bone marrow cells. Blood. 2000; 95: 581–585.