Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Feb 10.
Published in final edited form as: Nat Med. 2001 Sep;7(9):1035–1040. doi: 10.1038/nm0901-1035

Functional Small Diameter Neovessels using Endothelial Progenitor Cells Expanded Ex Vivo

Sunjay Kaushal 1,6, Gilad E Amiel 2, Kristine J Guleserian 1, Oz M Shapira 3, Tjorvi Perry 1, Fraser W Sutherland 1, Elena Rabkin 4, Adrian M Moran 5, Frederick J Schoen 4, Anthony Atala 2, Shay Soker 2, Joyce Bischoff 6, John E Mayer Jr 1
PMCID: PMC2818999  NIHMSID: NIHMS168398  PMID: 11533707

Abstract

Arterial conduits are increasingly preferred for surgical bypass because of inherent functional properties conferred by arterial endothelial cells, especially nitric oxide production in response to physiologic stimuli. We tested whether endothelial progenitor cells can replace arterial endothelial cells and promote patency of tissue-engineered small diameter blood vessels (4 mm). To address this, endothelial progenitor cells were isolated non-invasively from peripheral blood of sheep, expanded ex vivo, and subsequently seeded onto decellularized porcine iliac vessels. Endothelial progenitor cell-seeded grafts remained patent for 130 days as a carotid interposition graft in sheep, while non-seeded grafts occluded within 15 days. Importantly, the endothelial progenitor cell-explanted grafts exhibited contractile activity and nitric oxide–mediated vascular relaxation that was similar to native carotid arteries but which has not been reported previously for tissue-engineered small diameter grafts. These results suggest that endothelial progenitor cells can function similarly to arterial endothelial cells and thereby confer longer vascular graft survival. Due to their unique properties, endothelial progenitor cells may have more general applications for other tissue-engineered structures and also in treating vascular diseases.

Introduction

Cardiovascular disease, including coronary artery and peripheral vascular disease, is the leading cause of mortality in the United States 1. Surgical replacement or bypass surgery is the most common intervention for coronary and peripheral atherosclerotic diseases with 550,000 bypass cases performed per year 1. Autologous vessels are the preferred replacement grafts for diseased segments smaller than 5 mm in diameter, but many patients do not have suitable vessels2. An alternative approach for small diameter vessels is to tissue-engineer vessels by using scaffolds that are based either on a biodegradable or on a decelluarized matrix. Previously, in vitro culture of smooth muscle and endothelial cells on biodegradable or biological scaffold material produced a cellularized vessel with strong mechanical properties, but these grafts thrombosed shortly after implantation in vivo 3-5. An alternative approach for small diameter vessels is to allow repopulation of the tissue-engineered graft with host cells in vivo. A decellularized matrix, based either on intestinal collagen or native vessels, remodeled in vivo with infiltration of host smooth muscle cells and endothelial cells into the matrix and demonstrated good patency in vivo 6-8. One limitation of these studies is that the re-populated endothelial cells were not functional in endothelial-dependent relaxation assays performed on the explanted grafts, which may effect long-term patency rates. Furthermore, these results are difficult to extrapolate to humans because endothelialization of vascular grafts occurs more readily in the canine and rabbit models used in the studies9.

Endothelial cells are a crucial component of the normal vascular wall, providing an interface between the bloodstream and surrounding tissue of the blood vessel wall. Endothelial cells also play an important role in physiological events including angiogenesis, inflammation, and prevention of thrombosis. In addition to the endothelial cells that comprise the vasculature, there is evidence for non-hematologic, endothelial progenitor cells (EPCs) circulating in the blood 10-14. EPCs migrate to regions of the circulation that have pathologic conditions affecting the injured endothelium, including traumatic and ischemic injury 15-23. In normal adults, the concentration of EPCs in peripheral blood is 2-3 cells per milliliter, while levels are 3.5 fold higher in cord blood 14,23-24. Obtained from peripheral blood, EPCs provide a novel and attractive endothelial cell source yet to be explored for tissue-engineering cardiovascular structures that require a functional endothelium.

We propose that EPCs may be an ideal source of autologous endothelial cells for seeding small diameter grafts since no sacrifice of a native vessel is necessary to obtain the cells. By pre-seeding EPCs onto a vascular graft, a non-thrombogenic barrier for the vascular graft could be created, thereby improving the patency rate of the vascular graft in vivo. We tested the feasibility of this approach in a sheep model, an established model to study cardiovascular structures. In addition, we selected a decellularized graft since these grafts have well preserved extra-cellular matrix and mechanical properties resembling native vessels. The decellularized matrix thrombosed at high frequency when placed into the carotid position of sheep, consistent with previous vascular graft studies in the sheep and the need to have endothelial cells seeded on the graft prior to implantation 25.

Results

To test the feasibility of using EPCs to endothelialize tissue-engineered small diameter vascular grafts, venipuncture-derived blood samples (15cc) were obtained from one to two week old lambs without prior stimuli to mobilize EPCs from the bone marrow. EPCs were partially purified using a histopaque density gradient followed by plating the leukocyte fraction, which contained EPCs, onto fibronectin-coated plates as previously described12. Serial transfer of the cell suspension to new fibronectin-coated plates was performed to remove rapidly adherent hematopoietic cells. After 21 to 28 days in culture, the number of outgrowth colonies ranged between 2 to 3 per 1×107 total input cells. If allowed to continue growing, the outgrowth cells expanded exponentially and reached 1.6×109 cells by 3 weeks.

The outgrowth cells exhibited a cobblestone morphology that is characteristic of endothelial cells (Fig.1a). The endothelial phenotype was confirmed by immunostaining with specific antibodies directed against endothelial markers. A representative EPC preparation uniformly expressed CD31 on the cell-cell membrane borders (Fig.1b) and expressed von Willebrand factor (vWF) in cytoplasmic granules similar to Weibel-Palade bodies (Fig.1c). In addition, the EPCs incorporated acetylated low density lipoprotein (aLDL) (Fig.1d). E-selectin was minimally expressed in unstimulated cells (Fig.1e), but upon stimulation with lipopolysaccharide (LPS), was significantly upregulated in the EPCs (Fig.1f). To further document the endothelial phenotype, whole cell extracts were analyzed by immunoblot with specific antibodies to KDR/Flk-1, the vascular endothelial growth factor (VEGF) receptor-2 and Tie-2, the angiopoietin receptor (Fig.1g and h). The EPCs expressed equivalent amounts of Tie-2 and KDR when compared to human dermal microvascular endothelial cells (HDMEC). EPC response to VEGF and basic fibroblast growth factor (bFGF) was determined in cellular proliferation assays (Fig.1 i and j). Increasing VEGF and bFGF from 0 to 9 ng/ml resulted in a 4-fold and 8-fold induction of cell number, respectively. The proliferative effect of VEGF and bFGF was relatively similar to that seen with HDMEC when using the same experimental assay26. Consistent with the endothelial phenotype, the EPCs also formed capillary tubes within 6-12 hours when plated on Matrigel (data not shown). The expression of multiple endothelial markers detected by two different methods and the functional response to physiological stimuli unequivocally confirmed the endothelial identity of the outgrowth cells. Importantly for tissue-engineering, the endothelial phenotype remained constant for 20 passages, demonstrating the stability and feasibility of EPCs for clinical application.

Figure 1. EPCs from the peripheral blood exhibit the morphologic and phenotypic properties of endothelial cells.

Figure 1

Open in a new tab

EPCs exhibit typical cobblestone endothelial morphology (a), CD31 (b), von Willebrand factor (C) and incorporate aLDL (D). Unstimulated EPCs were negative for E-selectin expression (E), but LPS stimulated expression of E-selectin (f). g and h, Whole-cell extracts of two different EPC populations, EPC 9-19 and EPC 9-9B, HDMEC, and human fibroblasts were immunoblotted with antibodies shown on the left of each panel. Human fibroblast extracts were the negative control and HDMEC were the positive control. α-tubulin served as a control for protein loading. The three arrows indicate the fully glycosylated mature form of KDR (230 kDa), the partially glycosylated form of KDR (200 kDa), and unglycosylated form of KDR (150 kDa) 40. I and j, Proliferation of EPCs in response to increasing concentrations of VEGF and bFGF. Relative cell number was determined after 72 hours of incubation with the growth factors.

Once a consistent procedure for culturing EPCs from peripheral blood of sheep (n=20) was established, we seeded them onto vascular grafts. We choose decellularized vascular vessels that were prepared from 4-mm diameter porcine iliac arteries of 4 to 5 cm in length. After harvesting, the iliac vessels were incubated with trypsin, and then processed by a Triton X-100 detergent and ammonium hydroxide extraction technique (G. E. A., et. al., manuscript in preparation). The resultant vessel graft appeared devoid of cellular content, as seen by histological analysis (Fig.2a), by the lack of RNA detected in a reverse transcription-polymerase chain reaction assay (data not shown), and by the lack of an immune response after implantation of 3mm2 pieces of decellularized matrix into the peritoneal cavity of mice (data not shown). Furthermore, Movat staining demonstrated that the graft had a well-preserved extracellular matrix architecture, including internal and external elastic lamella (Fig.2b). After seeding a confluent monolayer of EPCs on the lumen of the graft, we next optimized preconditioning with variations in imposed shear stress to yield the maximal retention of EPCs prior to graft implantation in vivo. Previous studies demonstrate that up to 80% of non-preconditioned endothelial cells on ePTFE grafts are dislodged after exposure to shear stress in vitro or after implantation in vivo, and underscore the importance of pre-conditioning the endothelial cells on vascular grafts27-28. We simulated in vivo arterial conditions by subjecting the graft to high shear stress (25 dynes/cm2). Pretreatment of the seeded graft under conditions of low shear stress (1 dyne/cm2) for two days resulted in a cell density of 30 cells/mm (Fig.2c, open bar left). Subjecting the graft to high shear stress (25 dynes/cm2) for another two days resulted in a cell density of 13 cells/mm, a reduction of nearly 50% (Fig.2c, black bar left). In contrast, gradually increasing the shear stress from low (1 dyne/cm2) to high (25 dynes/cm2) over two days (Fig.2c, open bar right) followed by two additional days of high shear stress resulted in the same cell density of about 32 cells/mm (Fig.2c, black bar right). Scanning electron microscopy revealed a confluent monolayer of endothelial cells (Fig.2d).

Figure 2. Pre-implantation studies with EPCs seeded on the decellularized matrix.

Figure 2

Open in a new tab

Porcine iliac vessels were harvested and decellularized as described in Methods. a, Histology of the decellularized graft by hematoxylin and eosin stain (magnification 20X).b, Movat stain showed the internal and external elastin layers (dark brown) (magnification 20X).c, EPC density on the luminal surface of the decellularized grafts treated with low (1 dynes/cm2) or gradual (1 to 25 dynes/cm2) shear stress for 48-hours (open bars) and followed by 48 hour treatment with high (25 dynes/cm2) shear stress (black bars). d, Scanning electron microscopy of a EPC-seeded graft treated with gradual increases in shear stress for 48 hours and then an additional 48 hours of high shear stress showed a confluent layer of endothelial cells (magnification 880X, bar = 10 μ). e, The EPC-seeded graft induced NO production. In an organ chamber system, endothelial denuded guinea pig aortic ring was maximally contracted in the presence of U46619 and then relaxation was measured in the presence of increasing concentrations of calcium ionophore A23187 that was perfused directly into the EPC-seeded decellularized graft in the same organ chamber (▲ ). The relaxation effect mediated by A23187 was also measured in the presence of 10−3 M of L-NAME, an endothelial NO-synthase inhibitor (■ ). Relaxation was also recorded in the presence of increasing concentrations of SNP, an endothelial-independent NO donor ( ●). Data represent percent reduction in contraction and represent mean ± SEM derived from 4 grafts. *P<0.001 for dose response of A23187 vs. A23187 and L-NAME.

Once the pre-conditioning of the seeded graft was optimized, we assessed function of the EPCs by testing their ability to produce nitric oxide (NO). We assayed for NO because of its central role in normal vascular tissue homeostasis29-35. NO is not only a strong vasodilator but also an inhibitor of platelet aggregation, vascular smooth muscle proliferation, and leukocyte adhesion, all of which are strong anti-thrombotic and anti-atherogenic properties 29-35. We tested NO production by the EPC in a bio-assay using organ chamber methodology (Fig.2e). Endothelial-denuded segments of guinea pig aorta were contracted with U46619, a thromboxane analog, and then exposed to increasing concentrations of the calcium ionophore A23187 that was perfused through the EPC-seeded graft in the same organ chamber. Dose-dependent relaxation of the aorta segments was observed (Fig.2e, ▲). This dose-dependent relaxation was significantly attenuated by adding 10−3 M of the endothelial NO-synthase inhibitor L-NAME (77.5 % vs. 58.5 %, P<.001, Fig.2e, ■ ). When A23187 was perfused through the grafts that were not seeded with EPCs, the aortic segments remained fully contracted (data not shown). Exposing the aortic segments to sodium nitroprusside (SNP), an endothelial-independent NO donor, resulted in a dose-dependent relaxation, indicating normal smooth muscle function in the guinea pig aortic aortic segments (Fig.2e, Inline graphic). Taken together, these results showed that after pre-conditioning, the EPCs on the graft produced NO.

To assess the efficacy of EPC-seeded grafts in vivo, we performed implantation studies in sheep. Autologous EPCs were isolated from one to two week old lambs, expanded in culture, and seeded onto a decellularized grafts. The grafts were then preconditioned over two days, gradually increasing shear stress from low to high. Seeded (n=7) and unseeded (n=4) grafts, 4 to 5 cm in length, were implanted to replace a segment of the carotid artery using two end-to-end anastomoses. The only anti-coagulant administered was daily aspirin (81 mg) for 7 days post-operatively. All grafts demonstrated unobstructed flow at the time of implantation. We performed ultrasound Doppler flow studies daily and carotid duplex studies two weeks postoperatively to assess patency, flow, and thrombus formation. Three out of four control, non-seeded grafts occluded from thrombus formation within 5 days of implantation (Fig.3). One control graft was still patent when harvested at 15 days, but 50% of the lumen was obstructed by thrombus upon carotid duplex imaging and histologic analysis. All the grafts seeded with the EPCs were fully patent at the time of harvest, either 15 or 130 days after implantation (Fig.3). A representative arteriogram of a 130-day EPC-seeded explant demonstrated a smooth luminal surface with no signs of thrombus, obstruction, or aneurysmal dilation (Fig.4a). A macroscopic view of the 130 day EPC-seeded explant shown in Fig. 4e confirms the the smooth surface of the vessel lumen.

Figure 3. Long term patency of vascular grafts seeded with EPCs (---) or control vascular grafts with no endothelial cells (—).

Figure 3

Open in a new tab

Patency was defined by Doppler flow studies and confirmed by carotid duplex studies. The EPC-seeded grafts were harvested at 15 days (n=3) or at 130 days (n=4). The non-seeded grafts (controls) were harvested at 15 days (n=4).

Figure 4. Post-implantation studies of the vascular grafts.

Figure 4

Open in a new tab

a, Arteriogram of the EPC-seeded vascular graft 130 days after implantation. The location of the graft on the arteriogram was determined by staples at the sites of the anastomoses and is indicated by the two arrows. b, Seeded graft explanted at 15 days, showing a poorly organized thrombotic deposit overlying acellular graft media (asterisk). c-f, Seeded graft explanted at 130 days. c, Cellular intimal thickening as pannus from the adjacent arterial segments, infiltration of cells into the graft media, and confluent flattened cell monolayer lining the luminal surface of the graft (arrow). d, Section stained for extracellular matrix demonstrates the preservation of the original anatomic features, especially elastin (black). The internal elastic membrane is denoted by a white arrow. e, A macroscopic view of the explanted 130 day EPC-seeded vessel, with 1mm scale below. f, Staining of the cells within the intimal layer (i and inset) for smooth muscle alpha-actin. The medial layer (m) is unstained for alpha-actin. g, Staining of the cells overlying the intimal layer for vWF, denoted by arrow. Magnification in B, C, and F inset = 200X, in D and F= 100X, and in G = 400X. h, The pre-implant graft seeded with endothelial cells labeled with a fluorescent tracer. i, The 15 day and j, 130 explanted grafts showed the presence of CM-DiI-labeled endothelial cells on the luminal surface of the graft. Panels g-i are oriented with the lumen to the left.

Histological examination of the seeded grafts at 15 days revealed a confluent monolayer of endothelial cells (data not shown), however there were minimal areas of the lumen that had mild thrombus with cells concentrated at the thrombus-graft interface (Fig. 4b). The graft media was acellular. Histological examination of the entire length of the seeded grafts explanted at 130 days showed cellular intimal thickening as possible pannus from the adjacent arterial segments, infiltration of cells into the graft media, minimal adventitial inflammatory response, and a confluent flattened cell monoloayer lining the lumenal surface of the graft (Fig. 4c). Staining with the Movat stain demonstrated preservation of the original anatomic features of extracellular matrix, especially elastin (Fig. 4d). The cells within the intimal layer stained positively for smooth muscle alpha-actin, a characteristic of smooth muscle cells (Fig. 4f), while those lining the luminal surface stained for vWF, a characteristic of endothelial cells (Fig 4g).

To determine if the seeded EPCs were still present on the graft at 130 days, we labeled the EPCs prior to seeding with a fluorescent lipophilic carbocyanine tracer that incorporates into cell membranes. Examination of graft tissue sections prior to implantation demonstrated a confluently labeled endothelial cell layer (Fig.4h). Examination of the explanted grafts showed the presence of fluorescent EPC cells that covered approximately 80% and 10% of the graft lumen, at 15 and 130 days respectively (Fig.4i and j). The low percentage of fluorescently labeled EPCs at 130 days may be due to dilution of the membrane-incorporated dye with every cell division, which would suggest a high proliferation rate of the EPCs on the graft. Alternatively, some of the fluorescently labeled EPCs may have detached from the graft in vivo or the tracer dye may have diminished fluorescence at 130 days in vivo.

Since smooth muscle cells were present in the grafts, we evaluated the explanted graft for contraction and relaxation reactivity (Fig.5). The mid-segments of the 130 day explanted grafts were suspended in an organ chamber and contractility was measured in response to norepinephrine and serotonin. The 130 day explanted grafts contracted in a dose-dependent manner to norepinephrine and serotonin with a maximal response shown in Fig.5a. In comparison to the explanted grafts, matched native carotid segments generated two-fold more force in response to serotonin and eight-fold more force contraction in response to norepinephrine. These results indicate that the vascular smooth muscle cells observed in the grafts are indeed functional and respond to two physiologically important vasoconstrictors.

Figure 5. Vasomotor responsiveness of explanted grafts at 130 days.

Figure 5

Open in a new tab

a, Three mid-portion segments of each of the 130-day explant graft were tested for contractility activity in an organ chamber. Increasing doses of norepinephrine and serotonin, ranging from 10−9 M to 10−4 M, contracted the grafts and the maximal responses are shown at 10−4 M concentration. b, Matched segments of EPC-seeded grafts (◆), carotid artery (■), and saphenous vein ( ▲) were contracted with 10−4 M of serotonin and relaxation was assessed in response to increasing concentrations of acetylcholine. In addition, the EPC-seeded graft was relaxed in the presence of of acetylcholine and 10−3 M of L-NAME ( x ). *P<0.001 for EPC-seeded graft vs the L-NAME attenuated response and saphenous vein and P=.088 for EPC-seeded graft vs carotid artery. c, Matched segments similar to (b) were also assessed for relaxation in the presence of SNP. Data represent percent reduction in contraction and represent mean + SEM (n=12).

Increased endothelial NO production is associated with higher patency rates of coronary artery bypass arterial conduits, for instance radial artery or internal mammary artery 36-37. In a previous set of experiments we demonstrated the capacity of the EPC-seeded graft to produce NO in vitro (Fig. 2e). To further assess the endothelial function of EPC-seeded grafts in an attempt to evaluate their long-term patency, we compared the endothelial-dependent relaxation of explanted 130 day EPC-seeded grafts (◆) to that of native sheep carotid artery ( ■) and saphenous vein ( ▲). Segments of each vessel were pre-contracted with serotonin and exposed to increasing doses of acetylcholine inducing receptor-mediated NO production. Acetylcholine induced a dose-dependent relaxation in all three vessels (Fig.5b). Maximal relaxation of the EPC-seeded graft to acetylcholine was significantly greater than saphenous vein segments (68% ± 10% vs 22 ±6%, P<.001), but similar to carotid artery (68% ± 10% vs 81 ± 8%, P=.088) (Fig.5b). The acetylcholine induced dose-dependent relaxation of the explanted grafts was attenuated in the presence of the endothelial NO-synthase inhibitor L-NAME (68% ± 10% vs 22 ±6%, P<.001) (Fig.5b, x ). The three vessels responded equally to SNP (P=NS), an endothelial-independent NO donor, and achieved maximal relaxation, indicating normal vascular smooth muscle cell function of the explanted 130 day EPC-seeded grafts (Fig.5c). Taken together, these results suggest that the endothelial-dependent and endothelial-independent vasomotor properties of the EPC-seeded grafts are very similar to that of native carotid arteries, and, therefore, potentially may have long-term patency rates.

Discussion

Our results demonstrate that EPCs can be isolated from peripheral blood, expanded ex vivo, and retained on a decellularized graft. Upon implantation in vivo, the EPCs provide an autologous, non-thrombogenic luminal surface that maintained patency for at least 130 days. Furthermore, the EPC-seeded graft remodeled into a neovessel with vasomotor activity to physiological stimuli.

The amount of NO production correlates strongly with patency rates of various vessel conduits used in bypass surgery 36-37. For instance, the internal mammary artery has greater endothelial-dependent relaxation mediated by NO than the saphenous vein, which may play a role in higher patency rates of internal mammary artery grafts 36. We observed an endothelium-dependent relaxation of the 130 day explanted EPC-seeded grafts, which was comparable with native ovine carotid artery, but different from saphenous vein. We believe that this relaxation is dependent on NO production because the addition of the nitric oxide synthase inhibitor L-NAME attenuated the relaxation. Similar studies using the same methodology have shown the internal mammary artery and radial artery to have enhanced NO-mediated relaxation compared to saphenous vein, which may contribute to the long term patency of the internal mammary artery and radial artery grafts 37. Furthermore, since some of the seeded EPCs were still present on the graft after 130 days in vivo, EPCs may have intrinsic physiological characteristics, including NO production that may explain the patency rate of the EPC-seeded graft. This type of NO-mediated vascular relaxation suggests an important advantage of using EPCs in that this response has not been demonstrated previously in tissue engineered blood vessels.

Further studies are needed to identify the source of the smooth muscle cells in the neo-intima of the vascular graft. One likely source is smooth muscle cell migration across the anastomosis site from the adjacent artery. Examination of sections from the anasotomisis site suggests that smooth muscle cells may infiltrate the graft. Another potential source may involve transdifferentiation of the EPCs into smooth muscle cells. TGFβ1 induces the EPCs to express smooth muscle α–actin in vitro (S.K. and J.B., unpublished data) suggesting EPCs are more plastic than mature, differentiated endothelium and may have more potential for remodeling the graft in vivo. A third potential source may involve bone-marrow derived mesenchymal stem cells which have the potential to differentiate into smooth muscle-like cells and may be recruited into the graft from the circulation 38. Future studies are required to understand the vascular biology of the neo-intima formation on the decellularized graft in vivo.

The data presented here further demonstrates the importance of endothelializing vascular grafts for long-term patency. In the past, much effort has focused on endothelializing ePTFE grafts to increase their long-term patency in humans 39. Previously, the endothelial source for seeding grafts has relied on obtaining a biopsy from an existing blood vessel. Since EPCs can be obtained non-invasively from relatively small amounts of the patient’s peripheral blood, they may eliminate the need to sacrifice healthy vessels for harvesting autologous endothelial cells. In addition, EPCs may have broad clinical applications for tissue-engineering structures that may require an endothelial lining, such as heart valves. Another application of the EPCs may involve engineering the cells ex vivo to express specific genes for medical conditions and re-introducing them into selected areas of the body, for instance ischemic myocardium. We have already shown that EPCs are easily infected with a retrovirus and stably express the transgene in vitro (S. Kaushal, unpublished observations). Therefore, EPCs may ultimately provide a wide variety of therapeutic options for patients.

In summary, the EPC-seeded grafts develop into neovessels in vivo by 130 days and exhibited NO-mediated vascular relaxation. This study is limited by the size of the study groups and the limited number of intervals that were examined in vivo. Future long-term studies will determine whether aneurysmal dilation, thrombus formation, hyperplasia of the neo-intima, or other processes might impair the function of the neovessel. Our combined approach of seeding EPCs on a decellularized graft or other vascular grafts may provide an attractive approach for tissue engineering small diameter blood vessels. In addition, due to their physiological properties and easy accessibility, EPCs may have broad applications for treating vascular diseases.

Acknowledgements

We thank Dmitri Zvagelsky for technical assistance with preparing tissue sections, Ed Rhee for providing the arteriograms, Sulzer Carbomedics for providing the vessel bioreactor, and Kristin Gullage and Joey Fox for preparation of the figures. We also thank Drs. Judah Folkman, Charles Lowenstein, Stephen Roth, and Jay Schneider for critically reading the manuscript. This work was supported by RO1 HL 60490-03 and RO1 HL 60463-03 from the National Heart Lung and Blood Institute. The Sarnoff Foundation supported S.K.

Appendix

Methods

Isolation of EPCs from peripheral blood

EPCs were harvested from the internal jugular vein of 1 to 2 week old sheep as described1. After the initial 5 cc of blood was discarded, 15 cc of blood was mixed with 100 units of heparin. The leukoctye fraction was obtained by centrifuging on a histopaque density gradient (Sigma, St. Louis, MO) for 30 minutes at 1000 x g using Accuspin tubes (Sigma, St. Louis, MO). The cell pellet was resuspended in EBM-2 medium (Clonetics, San Diego, CA) with 20% fetal calf serum (Hyclone, Logan, UT) but without hydrocortisone and plated on fibronectin-coated plates. At 24-hour intervals for the next three days, the medium was transferred to new fibronectin-coated plates and thereafter the media was changed every 3 days. An aliquot of cells in the medium on day 3 were counted to determine input cells for final plating.

Indirect Immunoflourence

Indirect immunoflourence was performed as previously described using antibodies to CD31 (Santa Cruz Biotechnology, Santa Cruz, CA), vWF (Dako, Carpenteria, CA), and E-selectin26, and imaged for incorporation with fluorescent aLDL (Biomedical Technologies, Inc., Stoughton, MA). Cells were analyzed using a Zeiss Axiphot II fluorescence microscope and photographed using Ecktachrome p1600 color slide film.

Western Blot Analysis

Cell lysates were prepared as described41 and separated by SDS-PAGE and transferred to Immobilon-P membrane. Anti-KDR and anti-Tie-2 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) were used to detect these endothelial specific proteins.

Decellularized graft preparation

Freshly harvested porcine iliac blood vessels were placed in distilled water for one hour to remove blood elements. The vessels were then placed into a decellularizing solution containing 1% Triton X-100 and 0.1% ammonium hydroxide (Sigma, St. Louis, MO) and shaken (120-RPM) at 4°C for 72 hours. The solution was changed every 24 hours. After cleaning, the vessels were lyophilized and cold gas sterilized.

Isometric Studies

All isometeric tension experiments were performed as previously described37. Guinea pig aortic rings were suspended between 2 tungsten stirrups in an organ chamber bath of oxygenated 37°C Krebs solution. The vessels were increasingly stretched by 0.1 g increments to the optimal resting tension that produced a maximal response to 80 mmol/L KCl and allowed to relax for one hour. EPC-seeded grafts were directly attached to the aortic ring. After U46619 contraction of the aortic ring, relaxation of the aortic ring was induced by stimulating NO production in the EPCs with increasing concentrations of Ca+2 ionophore A23187. In addition, the graft response to A23187 and L-NAME at 10−3 M was assessed and also in the presence of increasing concentrations of SNP. The EPC-seeded 130-day explant was also tested for vasomotor activity using the same experimental design by sectioning the explanted grafts into 5mm rings. Three rings were tested from each of the four 130-day explants. Response to increasing concentrations of norepinephrine and serotonin was evaluated in the grafts. After maximal contraction with 10−4 M of serotonin, relaxation responses to increasing concentrations of acetylcholine and SNP were determined in the grafts. Matched segments of ovine carotid artery and saphenous vein were used as the controls.

EPC labeling

EPCs were labeled with 1μg/ml of cell tracker CM-DiI (Molecular Probes, Eugene, OR), a fluorescent carbocyanine dye, for 20 min. at 37° and then washed three times with PBS. Subsequently, the labeled EPCs were seeded on the graft.

Pre-conditioning of EPCs on the graft

The grafts were seeded at the density of 1×105 cells/cm2 in a rotating dynamic seeder for 6 hours at 12 rpm. EPC-seeded acellular grafts were then placed in a laminar flow bioreactor and shear stress was calculated using Poiseuille’s equation: τ=4ηQ/πr3, where τ=shear stress, η=fluid viscosity, Q=medium flow rate, and r=radius of the vessel. Shear stress was initiated either at 1 dyne/cm2 for two days or gradually increased from 1 dyne/cm2 to 25 dyne/cm2 by increasing 3 dyne/cm2 every 4 hrs over a 2 day period. Subsequently, all the grafts were subjected to 25 dyne/cm2 for another two days. Three mid-portion segments of each graft were analyzed for confluency by examining 5μm histological sections and counting the number of nuclei per mm around the circumference at the luminal surface42.

Statistical Analysis

All NO results are expressed as mean ± standard error of the mean (SEM). Statistical significance was evaluated using the unpaired Student’s t-test for comparisons between two means.

Surgical Implantation of the grafts

Anesthetized sheep (30 ± 3 kg) underwent a common carotid artery exposure through a longitudinal incision in the mid neck. Heparin (100 IU/kg; Elkins-Sinn, Cherry Hill, NJ) was administered intravenously prior to arterial clamping, and the graft was placed as an end-to-end anastomosis to the common carotid artery using a 6.0 prolene suture (Ethicon, Somerville, NJ). Arterial flow was re-established and the closure was sutured by layers. Animal care and handling complied with the Guide for the Care and Use of Laboratory Animals published by National Institutes of Health.

Explant Characterization

Tissue sections from the proximal and distal anastomoses and through the midportion of the graft were stained with hematoxylin and eosin. Smooth muscle cells and endothelial cells were identified by immunochemical analysis with antibodies to smooth muscle α-actin (Sigma, St. Louis, MO) and vWF (Dako, Carpenteria, CA), respectively

References

  • 1.Heart and Stroke Facts: Statistical Supplement. American Heart Association; 2000. http://www.americanheart.org/statistics. [Google Scholar]
  • 2.Tu JV, et al. Use of cardiac procedures and outcomes in elderly patients with myocardial infarction in the United States and Cannada. N Engl J Med. 1997;336:1500. doi: 10.1056/NEJM199705223362106. [DOI] [PubMed] [Google Scholar]
  • 3.Weinberg CB, Bell E. A blood vessel model constructed from collagen and cultured vascular cells. Science. 1986;213:397. doi: 10.1126/science.2934816. [DOI] [PubMed] [Google Scholar]
  • 4.L’Heureux N, Paquet S, Labbe R, Germain L, Auger FA. A completely biological tissue-engineered human blood vessel. FASEB J. 1998;12:47. doi: 10.1096/fasebj.12.1.47. [DOI] [PubMed] [Google Scholar]
  • 5.Niklason LE, et al. Functional arteries grown in vitro. Science. 1999;284:489. doi: 10.1126/science.284.5413.489. [DOI] [PubMed] [Google Scholar]
  • 6.Huynh T, et al. Remodeling of an acellular collagen graft into a physiologically responsive neovessel. Nature Biot. 1999;17:1083. doi: 10.1038/15062. [DOI] [PubMed] [Google Scholar]
  • 7.Schmidt CE, Baier JE. Acellular vascular tissues: natural biomaterials for tissue repair and tissue engineering. Biomaterials. 2000;21:2215. doi: 10.1016/s0142-9612(00)00148-4. [DOI] [PubMed] [Google Scholar]
  • 8.Sullivan SJ, Brockbank KGM. In: Principles of Tissue-Engineering. Langer B, Vacanti J, editors. 447-454. Academic Press; New York: 2000. [Google Scholar]
  • 9.Davids L, Dower T, Zilla P. In: Tissue Engineering of Prosthetic Vascular Grafts. Zilla P, Griesler HP, editors. 3-45. R.G. Landes Co; Austin, TX: 1999. [Google Scholar]
  • 10.Asahara T, et al. Isolation of Putative Progenitor Endothelial Cells for Angiogenesis. Science. 1997;275:964. doi: 10.1126/science.275.5302.964. [DOI] [PubMed] [Google Scholar]
  • 11.Yin A, et al. AC133, a novel maker for human hematopoietic stem and progenitor cells. Blood. 1997;90:5002. [PubMed] [Google Scholar]
  • 12.Shi Q, et al. Evidence for circulating bone marrow-derived endothelial cells. Blood. 1998;92:362. [PubMed] [Google Scholar]
  • 13.Gehling UM, et al. In vitro differentiation of endothelial cells from AC133-positive progenitor cells. Blood. 2000;95:3106. [PubMed] [Google Scholar]
  • 14.Lin Y, Weisdorf DJ, Solovey A, Hebbel RR. Origins of circulating endothelial cells and endothelial outgrowth from blood. J.Clin Invest. 2000;105:71. doi: 10.1172/JCI8071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Takahashi T, et al. Ischemia- and cytokine-induced mobilization of bone marrow-derived endothelial progenitor cells for neovascularization. Nat. Med. 1999;5:434. doi: 10.1038/7434. [DOI] [PubMed] [Google Scholar]
  • 16.Asahara T, et al. Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in physiological and pathological neovascularization. Circul. Res. 1999;85:221. doi: 10.1161/01.res.85.3.221. [DOI] [PubMed] [Google Scholar]
  • 17.Asahara T, et al. VEGF contributes to postnatal neovascularization by mobilizing bone marrow-derived endothelial progenitor cells. EMBO J. 1999;18:3964. doi: 10.1093/emboj/18.14.3964. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Kalka C, et al. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization. PNAS. 2000;97:3422. doi: 10.1073/pnas.070046397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kalka C, et al. Vascular endothelial growth factor 165 gene transfer augments circulating endothelial progenitor cells in human subjects. Circul. Res. 2000;86:1198. doi: 10.1161/01.res.86.12.1198. [DOI] [PubMed] [Google Scholar]
  • 20.Kalka C, et al. VEGF gene transfer mobilizes endothelial progenitor cells in patients with inoperable coronary disease. Ann. Thorac. Surg. 2000;70:829. doi: 10.1016/s0003-4975(00)01633-7. [DOI] [PubMed] [Google Scholar]
  • 21.Bhattacharya V, et al. Enhanced endothelialization and microvessel formation in polyester grafts seeded with CD34+ bone marrow cells. Blood. 2000;95:581. [PubMed] [Google Scholar]
  • 22.Crosby JR, et al. Endothelial cells of hematopoietic origin make a significant contribution to adult blood vessel formation. Circul. Res. 2000;87:728. doi: 10.1161/01.res.87.9.728. [DOI] [PubMed] [Google Scholar]
  • 23.Murohara, et al. Transplanted cord blood-derived endothelial precursor cells augment post-natal neovascularization. J. Clin. Invest. 2000;105:1527. doi: 10.1172/JCI8296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Peichev M, et al. Expression of VEGFR-2 and AC133 by circulating human CD34+ cells identifies a population of functional endothelial precursors. Blood. 2000;95:952. [PubMed] [Google Scholar]
  • 25.Ortenwall P, Bylock A, Kjellstrom BT, Risberg B. Seeding of ePTFE carotid interposition grafts in sheep and dogs: Species-dependent results. Surgery. 1987;103:199. [PubMed] [Google Scholar]
  • 26.Kraling BM, Bischoff J. A simplified method for growth of human microvascular endothelial cells results in decreased senescence and continued responsiveness to cytokines and growth factors. In Vitro Cell. Dev. Biol. 1998;33:308. doi: 10.1007/s11626-998-0007-z. [DOI] [PubMed] [Google Scholar]
  • 27.Rosenman JE, Kempczinski RF, Pearce WH, Silberstein EB. Kinetics of endothelial cell seeding. J. Vasc. Surg. 1995;2:778. [PubMed] [Google Scholar]
  • 28.Miyata T, et al. Delayed exposure to pulsatile shear stress improves retention of human saphenous vein endothelial cells on seeded ePTFE grafts. J. Surg. Res. 1991;50:485. doi: 10.1016/0022-4804(91)90029-l. [DOI] [PubMed] [Google Scholar]
  • 29.Keaney JF, Jr., Vita JA. Atherosclerosis, oxidative stress, and antioxidant protection in endothelium-derived relaxing factor action. Prog Cardiovasc. Dis. 1995;38:129. doi: 10.1016/s0033-0620(05)80003-9. [DOI] [PubMed] [Google Scholar]
  • 30.Quyyumi AA, et al. Nitric oxide activity in the human coronary circulation. Impact of risk factors for coronary atherosclerosis. J. Clin. Invest. 1995;95:1747. doi: 10.1172/JCI117852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Azuma H, Ishikawa M, Sekizaki S. Endothelium-dependent inhibition of platelet aggregation. Br. J. Pharmacol. 1986;88:411. doi: 10.1111/j.1476-5381.1986.tb10218.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Radomski MW, Palmer RM, Moncada S. The role of nitric oxide and cGMP in platelet adhesion to vascular endothelium. Biochem. Biophys. Res. Commun. 1987;148:1482. doi: 10.1016/s0006-291x(87)80299-1. [DOI] [PubMed] [Google Scholar]
  • 33.Kubes P, Kurose I, Granger DN. NO donors prevent integrin-induced leukocyte adhesion but not P-selectin-dependent rolling in postischemic venules. Am. J. Physiol. 1994;267:H931. doi: 10.1152/ajpheart.1994.267.3.H931. [DOI] [PubMed] [Google Scholar]
  • 34.Marks DS, et al. Inhibition of neointimal proliferation in rabbits after vascular injury by a single treatment with a protein adduct of nitric oxide. J. Clin. Invest. 1995;96:2630. doi: 10.1172/JCI118328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Luscher TF, et al. Difference between endothelium-dependent relaxation in arterial and in venous coronary bypass grafts. N. Eng. J. Med. 1988;319:462. doi: 10.1056/NEJM198808253190802. [DOI] [PubMed] [Google Scholar]
  • 36.Pearson PJ, Evora PRB, Schaff HV. Bioassay of EDRF from internal mammary arteries: Implications for early and late bypass graft patency. J. Thorac. Surg. 1992;54:1078. doi: 10.1016/0003-4975(92)90073-d. [DOI] [PubMed] [Google Scholar]
  • 37.Shapira OM, et al. Enhanced nitric oxide-mediated vascular relaxation in radial artery compared with internal mammary artery or saphenous vein. Circulation. 1999;100:II–322. doi: 10.1161/01.cir.100.suppl_2.ii-322. [DOI] [PubMed] [Google Scholar]
  • 38.Shimizu K, et al. Host bone marrow cells are a source of donor intimal smooth muscle-like cells in the murine aortic transplant arteriopathy. Nat. Med. 2001;7:738–741. doi: 10.1038/89121. [DOI] [PubMed] [Google Scholar]
  • 39.Deutsch M, Meinhart J, Zilla P. In: Tissue Engineering of Prosthetic Vascular Grafts. Zilla P, Griesler HP, editors. R.G. Landes Co; Austin, TX: 1999. pp. 180–187. [Google Scholar]
  • 40.Shen B, et al. Homologous up-regulation of KDR/Flk-1 receptor expression by vascular endothelial growth factor in vitro. J. Biol. Chem. 1998;273:29979. doi: 10.1074/jbc.273.45.29979. [DOI] [PubMed] [Google Scholar]
  • 41.Shen B-Q, et al. Hepatocyte growth factor stimulates the differentiation of human tracheal epithelia in vitro. Am. J. Phys. 1997;272:L1115. doi: 10.1152/ajplung.1997.272.6.L1115. [DOI] [PubMed] [Google Scholar]
  • 42.Dardik A, Liu A, Ballermann BJ. Chronic in vivo shear stress stimulates endothelial cell retention on prosthetic vascular grafts and reduces subsequent in vivo neointimal thickness. J Vasc. Biol. 1999;29:157. doi: 10.1016/s0741-5214(99)70357-5. [DOI] [PubMed] [Google Scholar]

RESOURCES