tissue engineering of small-diameter vascular grafts by endothelial progenitor cells seeding...

Tissue engineering of small-diameter vascular grafts by endothelialprogenitor cells seeding heparin-coated decellularized scaffolds

Min Zhou,1 Zhao Liu,1 Cheng Liu,1 Xuefeng Jiang,2 Zhiqing Wei,1 Wei Qiao,1 Feng Ran,1

Wei Wang,1 Tong Qiao,1 Changjian Liu1

1Department of Vascular Surgery, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School,

Nanjing 210008, People’s Republic of China2College of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, People’s Republic of China

Received 4 October 2010; revised 22 May 2011; accepted 25 June 2011

Published online 24 November 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.31928

Abstract: Successful construction of a small-diameter bioarti-

ficial vascular graft remains a great challenge. This study

reports on novel tissue engineering vascular grafts (TEVGs)

constructed by endothelial progenitor cells and heparin-

coated decellularized vessels (DV). The DVs were fabricated

from canine carotid arteries with observable depletion of cel-

lular components. After heparin coating, the scaffolds pos-

sessed excellent antithrombogeneity. Canine endothelial

progenitor cells harvested from peripheral blood were

expanded and seeded onto heparin-coated DVs and cocultured

in a custom-made bioreactor to construct TEVGs. Thereafter,

the TEVGs were implanted into the carotid arteries of cell-

donor dogs. After 3 months of implantation, the luminal surfa-

ces of TEVGs exhibited complete endothelium regeneration,

however, only a few disorderly cells and thrombosis overlaid

the luminal surfaces of control DVs grafts, and immunofluo-

rescent staining showed that the seeded cells existed in the

luminal sides and the medial parts of the explanted TEVGs

and partially contributed to the endothelium formation. Specif-

ically, TEVGs exhibited significantly smaller hyperplastic neo-

intima area compared with the DVs, not only at midportion

(0.64 6 0.08 vs. 2.13 6 0.12 mm2, p < 0.001), but also at anas-

tomotic sites (proximal sites, 1.03 6 0.09 vs. 3.02 6 0.16 mm2,

p < 0.001; distal sites, 1.84 6 0.15 vs. 3.35 6 0.21 mm2, p <

0.001). Moreover, TEVGs had a significantly higher patency

rate than the DVs after 3 months of implantation (19/20 vs. 12/

20, p < 0.01). Overall, this study provided a new strategy to

develop small-diameter TEVGs with excellent biocompatibility

and high patency rate. VC 2011 Wiley Periodicals, Inc. J Biomed

Mater Res Part B: Appl Biomater 100B: 111–120, 2012.

Key Words: tissue engineering vascular graft, endothelial

progenitor cells, decellularized vessels, small-diameter, hepa-

rin coating

How to cite this article: Zhou M, Liu Z, Liu C, Jiang X, Wei Z, Qiao W, Ran F, Wang W, Qiao T, Liu C. 2012. Tissue engineering ofsmall-diameter vascular grafts by endothelial progenitor cells seeding heparin-coated decellularized scaffolds. J Biomed MaterRes Part B 2012:100B:111–120.

INTRODUCTION

Atherosclerosis and heart diseases are still the leadingcauses of morbidity and mortality worldwide. Therapies forcoronary artery and peripheral vascular diseases oftenrequire replacement of the damaged vessels with vasculargrafts. For reconstruction of large arteries, such as the aortaor iliac artery, the current commercial grafts made fromexpanded polytetraflouroethylene (ePTFE) or Dacron areutilized satisfactorily. However, synthetic grafts are not suit-able for reconstruction of smaller diameter (internal diame-ter < 5 mm) arteries, due to thrombosis, limited re-endo-thelialization and neointimal hyperplasia, mainly owing tothe inherent properties of the synthetic materials. Therefore,surgeons routinely use autologous vessels for such recon-struction procedures. Unfortunately, the number of appro-priate vessels is limited in many patients due to coexisting

diseases, size mismatch, or previous procedures. Althoughconsiderable research has focused on the development ofnovel small-diameter vascular grafts for decades, there isstill no adequate alternative to the autologous vessels.1

A recent promising approach for the small-diameter vas-cular grafts is the use of nature vascular scaffolds with totaldepletion of cellular antigens, which have shown goodmechanical properties and biocompatibility in vivo.2,3 It wasalso observed, however, the acellular luminal surface with-out endothelial cells (ECs) coverage carried a substantialrisk for thrombosis when directly exposed to the blood,which would finally effect the long-term patency rate.4,5 Forthat reason, most studies focused on the creation of tissueengineering vascular grafts (TEVGs) by recellularizing scaf-folds with host vascular cells prior to implantation.6–9

In the past 20 years, clinical studies indicated that the

Additional Supporting Information may be found in the online version of this article.

Correspondence to: T. Qiao; e-mail: [email protected]

Contract grant sponsor: Important Science Project of Jiangsu Health; contract grant number: K200609

VC 2011 WILEY PERIODICALS, INC. 111

ECs-seeded grafts had high patency rates in human cardiacartery10 and lower extremity artery bypass grafting.11

Despite such successful results, the clinical application of theECs-seeded graft was still hampered by the lack of a conven-ient source of ECs.12 Fortunately, recent articles reported thatendothelial progenitor cells (EPCs) were expected to havemuch higher proliferative potential than mature ECs,13,14 andthe procedure of using EPCs was much less invasive than thesurgical harvest of ECs from large veins or tissue microves-sels.15 Therefore, it was anticipated that EPCs harvested fromperipheral blood mononuclear cells (PBMCs) and proliferatedeffectively in vitro, might serve as a very promising cellsource of vascular tissue engineering.

In this study, we attempted to develop small-diameterTEVGs by seeding EPCs onto the heparin-coated decellular-ized scaffolds. For that purpose, EPCs were harvested fromcanine PBMCs and expanded in vitro, the endothelial lineagephenotype of the EPCs was identified by immunofluorescentstaining. And heparin-coated decellularized vessels (HDVs)were fabricated and served as vascular tissue scaffolds.Then TEVGs were constructed in vitro and implanted incell-donor canine models for 3 months, the TEVGs regenera-tion and patency rate were examined, which would eventu-ally be needed to demonstrate its clinical applicability.

MATERIALS AND METHODS

Preparation and characterization of decellularizedvesselsThe protocol for the care and use of animals was conductedaccording to the Guide for the Care and Use of LaboratoryAnimals approved by the Ethical Committee of Researchesof Nanjing University.

Decellularized vessels (DVs) were prepared as previ-ously described.6 In brief, freshly harvested canine carotidarteries were placed into a solution containing 1% TritonX-100 (Sigma) in phosphate buffered saline (PBS) for 24 h,then placed in a solution of 0.1% trypsin (Biochrom) with0.02% ethylenediaminetetraacetic acid in PBS for another24 h. This procedure was followed by an incubation withribonuclease (RNase) A (20 lg/ml) (Sigma) and desoxyribo-nuclease (DNase) (0.2 mg/ml) (Sigma) for 2 h. All stepswere conducted in a 5% CO2/95% air atmosphere at 37�Cunder continuous shaking. After cleaning, the grafts were ly-ophilized and sterilized in ethylene oxide gas.

To evaluate the effect of decellularization, a segmentfrom the scaffolds were fixed in 10% neutral buffered for-malin, embedded in paraffin, and sliced for hematoxylin-eo-sin (H&E) staining. Sliced samples were also incubated withhorseradish peroxidase-conjugated anti-Mouse major histo-compatibility complex class I (MHC I) antibody (Biomeda),and followed by visualization with 3, 30-diaminobenzidine.The luminal surface of the DVs was also analyzed by scan-ning electron microscopy (SEM) as follows: specimens werefixed in 1% buffered glutaraldehyde and 0.1% bufferedformaldehyde for 1 and 24 h, respectively, then dehydratedwith a graded ethanol series, critical point dried, sputtercoated with platinum, and observed by SEM (ISI-SX-40, Aka-shi, Tokyo, Japan).

Heparin coatingCovalent immobilization of heparin to DVs was performedusing 1-ethyl-3- (3-dimethyl aminopropyl) carbodiimide(EDC) and N-hydroxysuccinimide (NHS) (both from PierceBiotechnology, Rockford, IL) according to Yao et al.16 Briefly,1 g heparin, 2 g EDC, and 1.2 g NHS were added into500 mL 2-morpholinoethanesulfonic acid buffer (0.05 M, pH5.6) for 10 min at 37�C to activate carboxylic acid groups ofheparin. Then DVs were immersed into the reagent solutionfor 4 h at 37�C under gentle shaking. Following heparinimmobilization, vessels were rinsed in 0.1 M Na2HPO4(2 h), 4 M NaCl (four times for 24 h), and distilled water(three times for 24 h).

Characterization of coating heparin via plateletadhesion experimentTo characterize the effect of coating heparin, platelet adhe-sion experiment was performed as previously reported.17

Briefly, HDVs and uncoated DVs were cut into disk shapes(5 mm in diameter) and placed into a 24-well tissue cultureplate. The PBS was allowed to stand in the wells for a dayto equilibrate the surfaces. Then the PBS was removed and1 ml of platelet-rich plasma (supplied by Blood Center ofNanjing Red Cross) was poured into each well and storedfor 3 h at 37�C. Samples were gently rinsed with PBS andtreated with 2.5% glutaraldehyde for 30 min at room tem-perature then cleaned and dehydrated by systemic immer-sion in a series of ethanol-water solutions for 30 min eachand allowed to evaporate at room temperature. The plate-let-attached surfaces were observed by SEM. The commer-cial ePTFE vascular grafts served as a reference.

Culture and characterization of EPCsCanine PBMCs were isolated by density-gradient centrifuga-tion of adult canine peripheral blood using Ficoll-PaquePlus (Amersham Biosciences, Uppsala, Sweden).18 The cellswere resuspended in microvascular growth medium-2(EGM-2MV; Cambrex, Walkersville, MD), which comprisedendothelial basal medium-2 and SingleQuots containing 5%fetal bovine serum, vascular endothelial growth factor,human fibroblast growth factor B, human epidermal growthfactor, insulin-like growth factor-1, ascorbic acid, hydrocorti-sone, and GA-1000. The PBMCs (5 � 105/well) were thenimmediately plated on fibronectin-coated 35-mm diametertissue culture plates (BD Biosciences, Bedford, MA). After 3days of culture, nonadherent cells were discarded, and freshmedium was applied. The attached cells were continuallycultured with complete EGM-2 medium over the course of 3weeks. Culture medium was changed every 3 days. Themorphological changes of adherent cells were visualizedwith Olympus phase-contrast microscopy over culture(Olympus Optical Co. Ltd, Tokyo, Japan).

To identify EC-like cells in cultures as endothelial lineagecells, the cellular expression of von Willebrand factor(vWF), vascular endothelial growth factor receptor-2 (Flk-1)and VE-cadherin were examined according to previousreport.19 Briefly, first-passage cells were fixed with 4%paraformaldehyde for 30 min at 37�C. As the primary

112 ZHOU ET AL. TISSUE ENGINEERING VASCULAR GRAFTS

antibodies, goat polyclonal anti-vWF (Abcam, UK), anti-Flk-1(Santa Cruz Biotechnology, Heidelberg, Germany), and anti-VE-cadherin (Transduction, Lexington, UK) were applied tocells. After the cells were washed with PBS, all the antigenswere detected by incubation in 1:500 dilution of Alexa Fluor546 rabbit anti-goat IgG conjugate (Molecular Probes) for 1hour at room temperature, after staining the samples wereobserved with a fluorescent microscope (Olympus). To con-firm the EPCs phenotype, the attached EC-like cells wereincubated with DiI-labeled acetylated LDL (DiI-acLDL; 10lg/ml; Molecular Probes) at 37 �C for 1 h, and then fixedwith 4% paraformaldehyde for 30 min at 37�C and incu-bated with fluorescein isothiocyanate-labeled Ulex europeusagglutin (FITC-UEA-I, 10 lg/ml; Sigma) for 4 h at 37�C, af-ter that, the samples were observed with a phase-contrastfluorescent microscope (Olympus). Cells demonstrating dou-ble-positive fluorescence were identified as differentiatingEPCs.20 Additionally, for analysis of capillary tube formation,detached EPCs were seeded onto 4-well plates (40,000cells/well) precoated with 300 uL matrigel (BD Bioscien-ces), and observed after 24 hours in culture.

Construction of TEVGsThe construction of TEVGs was divided into two procedures,static seeding and dynamic culture, respectively. For seedingprocedure, the cultured EPCs were harvested and diluted with1 ml of EGM-2 medium, and labeled with 1 lg/mL of a fluo-rescent cell tracker (CM-DiI; Molecular Probes, Eugene, OR) forin vivo cell tracing.8 Then the cell suspension (1.0 � 106 cells/ml, 0.5 ml) was added into HDV (length, 50 mm; inner diame-ter, 3 mm) and subsequently immersed in the culture mediafor 6 h at 37�C. Another cell suspension was then added inthe same manner and the graft was rotated 90� around itslongitudinal axis. After repeating this procedure for four times(total seeded cells, 2.0 � 106 cells) the graft was incubated foranother 24 h to ensure complete cells attachment.

After 2 days of incubation in EGM-2 culture mediumstatically, the EPCs-seeded graft was obtained and dynamicculture was followed as previously reported.12 In brief, thecell-seeded graft was connected to a circulatory loop system,which consisted of a roller pump (Zhisun Instrument, Shang-hai, China) upstream of the graft, and an outflow reservoirdownstream of the graft. The graft was tied to the circulatoryloop filled with culture medium, which flowed from the rollerpump through the graft and into the outflow reservoir.Medium in the outflow reservoir was pumped up with theroller pump and re-circulated. To simulate physiological flowrates, the pump setting was 70 strokes/ min and yielded aflow rate of 60 ml/min. At this flow rate, the calculated shearstress at the graft was 30 dyn/ cm2. The entire system was in-stalled in an incubator at 37�C in a humidified environmentwith 5% CO2. After 7 days of perfusion, the graft was retrievedfor SEM observation and animal implantation.

Surgical implantation and explant characterizationThe TEVGs (n ¼ 20) were implanted as carotid artery inter-position grafts in cell-donor dogs. Anesthesia was inducedwith injection of intramuscular ketamine (30 mg/kg) and

intravenous pentobarbital (30 mg/kg), and then maintainedwith isoflurane and oxygen. Through a longitudinal mid-neck incision, bilateral common carotid arteries wereexposed. Prior to arterial clamping, heparin (100 U/kg) wasadministered intravenously. The grafts were placed as anend-to-end anastomosis to the common carotid arteriesusing a 7-0 prolene suture. The nonseeded DVs grafts (n ¼20) were implanted in a similar manner in the contralateralsides as controls. The length of both graft segments wasapproximately 4–5 cm. The arterial flow was reestablishedand the closure was sutured by layers. All animals receivedaspirin (100 mg daily) postoperatively. The implanted graftpatency was monitored with a handle Doppler probe (HPSonos 4500, Philips, Forestville, CA) every month. After 3months of implantation, all the grafts were explanted forhistological examination. The mid-portion and anastomoticsites were stained with H&E and Masson’s trichrome stain-ing, and the luminal surface was analyzed with SEM. Thedegree of neointima formation was evaluated by histomorph-ometry using computer-assisted planimetry system (ImagePro Plus 3.0.1 software, Media Cybernetics, Bethesda, MD).For detection of CM-DiI-labeled cells and vWF-positive cells,the mid-portion tissue sections were stained immunofluores-cently for vWF using FITC-conjugated anti- goat IgG second-ary antibodies (Jackson ImmunoResearch Laboratories, WestGrove, PA) and analyzed using a confocal microscope(LSM510, Carl Zeiss, Oberkochen, Germany).

Statistical analysisQuantitative data were expressed as mean 6 SD. Pearson’schi-squared test was employed to compare the patencyrates between the TEVGs and DVs. In addition, unpairedStudent’s t test was conducted to compare the difference ofthe neointima area at two sites between TEVGs and DVs. Allanalyses were performed using SAS software (version 9.1,SAS Institute, Cary, NC). A p value of less than 0.05 was con-sidered to be statistically significant.

RESULTS

Characterization of decellularized vesselsDecellularized vascular grafts were prepared by removing cel-lular components and leaving the native extracellular matrix(ECM) of the arteries. H&E staining did not show any signs ofremaining nuclear materials in the vessel walls [Figure 1(A,B)].The immunostaining of MHC I confirmed the efficacy of thedecellularization process in removing the majority of cellularelements [Figure 1(C,D)]. SEM further verified the removal ofcellular debris from luminal surface, while the basic extracellu-lar microstructure remained intact [Figure 1(E,F)].

Platelet adhesion experimentPlatelet adhesion is an important test for the evaluation ofthe blood compatibility of the grafts. The result of this testwas observed by SEM in this study. It showed that plateletswere rarely observed on the surfaces of HDVs [Figure 2(A)],while DVs showed much higher platelet adhesion, most ofthe adhered platelets were spread and deformed and theymight be activated [Figure 2(B)].

ORIGINAL RESEARCH REPORT

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | JAN 2012 VOL 100B, ISSUE 1 113

Morphological observation and characterization of EPCsAfter 7 days of culture, cluster-like adherent cells were found,having assumed a spindle, triangular, polygonal, or irregularshape [Figure 3(A)]. These cell islets enlarged and changed toa layer of contact inhibited monolayer cells, and the cells wereconfluent by 3 weeks, showing the typical ‘‘cobblestone’’appearance characteristic of an EC monolayer [Figure 3(B)].The cells were also assembled into primitive vascular tube-likestructures when plated in Matrigel [Figure 3(C)]. After 3weeks of culture, nearly all of the cobblestone-like cells (>95%) showed double-positive results in the LDL-uptake assaysand the UEA-lectin binding tests [Figure 3(D–F)]. Additionally,the immunofluorescence studies showed the majority of theadherent cells stained positively for Flk-1 [Figure 3(G)], vWF[Figure 3(H)], and VE-cadherin [Figure 3(I)]. The cells couldbe passaged several times without significant alterations totheir morphology or growth characteristics.

Construction of TEVGsWhen highly proliferative EPCs were statically seeded intosmall-diameter HDVs, EC-like cells adhered tightly to the

lumenal surface of the grafts after 2-day culture. The cellsspread out fully and possessed a sustainable growth capa-bility [Figure 4(A,C)]. The seeded grafts were translocatedto the closed circulatory loop apparatus and cocultured for7 days, after that, EC-like cells elongated and aligned in the

FIGURE 1. Characterization of decellularized vessel. (A) H&E staining of

natural artery (original magnification, 50�, the scale bar indicates

200 lm). (B) H&E staining of DV showed complete removal of cellular

components (original magnification, �50, the scale bar indicates 200 lm).

(C) Masson staining of natural artery (original magnification, �50, the

scale bar indicates 200 lm). (D) Masson staining of DV showed wellpre-

served extracellular matrix (original magnification, �50, the scale bar indi-

cates 200 lm). (E) SEM of the luminal surface of natural artery (original

magnification, �1000, the scale bar indicates 10 lm). (F) SEM of the lumi-

nal surface of DV (original magnification,�1000, the scale bar indicates 10

lm). 49 � 64mm (500 � 500 DPI). [Color figure can be viewed in the online

issue, which is available at wileyonlinelibrary.com.]

FIGURE 2. Characterization of heparin-coated decellularized vessels.

SEM after platelet adhesion test exhibited less platelets adhering on

the heparin-coated decellularized graft compared with control grafts.

(A) Heparin-coated decellularized vessels. (B) Heparin uncoated decel-

lularized vessels. (C) ePTFE vascular graft. (The arrowheads indicate

adhered platelets, original magnification �1500, the scale bars indi-

cate 10 lm.) 49 � 36mm (500 � 500 DPI).


direction of flow, and adhered tightly to each other to main-tain the high integrity of confluent monolayer structure andalmost all areas were covered with a cell monolayer [Figure4(B,D)], similar to that of native artery.

Surgical implantation and explanted characterizationSegments of the canine common carotid arteries werereplaced by either TEVGs or DVs grafts. The animals wereperiodically investigated by Doppler ultrasound after im-plantation (Figure 5). Out of the 20 TEVGs, 19 maintainedpatency for up to 3 months, one occluded within the firstmonth due to thrombus formation. However, only 12 DVswere still patent when explanted at 3 months, the othereight were observed with thrombotic occlusion within thefirst month after implantation (Figure 6). The patency ratesbetween TEVGs and DVs differed significantly (p < 0.01).

The patent grafts were retrieved at 3 months for histo-logical analyses. The H&E staining showed TEVGs exhibitedendothelium regeneration on the luminal surface and adense population of fibroblast-like cells in the inner layers ofthe media [Figure 7(A)], but only a few poorly organizedcells and thrombotic deposit overlaid the luminal surfaces ofDVs, and massive disorderly cells were present throughoutthe reconstructed walls [Figure 7(B)]. SEM revealed that theluminal surfaces of both grafts were covered with a confluentmonolayer of cobblestone-like cells, but the orientation highlyparallel to the direction of arterial flow observed in TEVGs[Figure 7(C)] was not seen in the DVs [Figure 7(D)]. Masson’strichrome staining displayed that collagen and elastin werereconstructed during the 3 months in both grafts [Figure7(E,F)] and the subsequent morphometric analysis revealedthat the TEVGs (n ¼ 19) exhibited significantly smaller

FIGURE 3. Characterization of EPCs derived from canine peripheral blood. (A) The cluster-like adherent EPCs after 7 days proliferation (original

magnification, �100). (B) The cobblestone appearance of EPCs at 3 weeks (original magnification, �100). (C) Vascular tube formation on Matrigel

(original magnification, �20). (D) Binding of UEA-1 as green flurescence by the EPCs (original magnification, �100). (E) Uptake of Ac-LDL as red

flurescence from the medium by EPCs (original magnification, �100). (F) Merging of the flurescence imanges displayed in D and E (original

magnification, �100). (G) EPCs immunofluorescent staining positive for von Willebrand factor (original magnification, �100). (H) Immunofluores-

cent staining for flk-1 (original magnification, �100). (I) Immunofluorescent staining for VE-cadherin (original magnification, �100). 59 � 58mm

(300 � 300 DPI). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]



hyperplastic neointima area compared with the DVs (n ¼ 12),not only at midportion (0.64 6 0.08 vs. 2.13 6 0.12 mm2, p <

0.001), but also at anastomotic sites (proximal sites, 1.03 60.09 vs. 3.02 6 0.16 mm2, p < 0.001; distal sites, 1.84 6 0.15vs. 3.356 0.21 mm2, p < 0.001) [Figure 8].

Next, we investigated if the seeded EPCs survived afterimplantation. EPCs labeled with fluorescent carbocyaninedye (CM-DiI) prior to seeding were identified in the luminalsides and the medial parts of the explanted TEVGs retrievedat 3 months [Figure 9(A)]. Immunofluorescent staining forvWF showed that endothelium (green) formed extensively[Figure 9(B)] in TEVGs. The merged images of vWF andCM-DiI double-positive cells showed that endothelium (lightyellow, arrows) formed by seeded EPCs was present in theluminal surface of TEVGs [Figure 9(C)].

DISSCUSION

Current small-diameter commercial vascular grafts exhibit ahigh rate of failures due to thrombosis, calcification, infec-

tion, and no growth potential. Tissue engineering holds agreat promise in providing ideal vascular grafts and over-coming above fatal shortcomings.21 In this study, we demon-strated the feasibility of seeding autologous EPCs onto HDVsto construct novel TEVGs (See Supporting information).

The biomaterial scaffold plays a key role in most tissueengineering strategies, and the ideal scaffold should providea suitable environment for tissue development. Naturallyderived decellularized scaffold has the potential advantageof interacting with specific cells and is an essential compo-nent of the ECM that supports cell expansion.22 Hence, wefabricated DVs from native carotid arteries by detergent andenzymatic treatment, and histological analysis showed thecellular components were removed, while leaving theporous biological three-dimensional architecture intact,which could provide physiologically proper microenviron-ment for vascular cells repopulation.23 Immunohistochemi-cal analysis of MHC I, which was involved in immuneresponses against transplanted tissue and expressed in all

FIGURE 4. The morphology of highly proliferative EC-like cell-seeded vascular graft. (A) Fluorescent microscope images of EPCs labeled with

CM-DiI after 2 days of culture under static conditions (original magnification, �100; the scale bar indicates 100 lm). (B) Fluorescent microscope

images of EPCs labeled with CM-DiI after 7 days of dynamic culture (original magnification, �100; the scale bar indicates 100 lm). (C) SEM ob-

servation of lumenal surface of the graft after 2 days of culture under static conditions (original magnification, �500; the scale bar indicates 20

lm). (D) SEM observation of lumenal surface of the graft after 7 days of dynamic culture. EPCs were elongated and aligned in the direction of

flow, and maintained confluent cell-layer coverage on the lumenal surface of the vascular graft, flow is indicated by the arrow (original magnifi-

cation, �500; the scale bar indicates 20 lm). 49 � 44mm (500 � 500 DPI). [Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]


blood vessel cells, further confirmed the removal of the vas-cular cells. Quite a few studies showed that the mechanicalproperties such as burst strength, compliance, and sutureretention strength were not significantly different betweendecellularized tissue matrices and native vascular tissues.2–4

Others reported allogeneic DVs showed low immunogenicityin vivo24 and more resistance to infection than PTFE graftwhen used as hemodialysis access.25

Furthermore, we immobilized heparin to the DVs usingEDC and NHS as cross-link agents in present study. Duringthis process, the carboxyl groups on the heparin were acti-vated to succinimidyl esters, which reacted with aminofunctions on the collagen and elastin to zero length cross-links.25 This modification led to covalent linkage of heparinto the DVs, which unlike the ionic linkage of heparin usedby several other groups, appears to be more stable and longlasting.26 Thereafter, we applied platelet adhesion test todemonstrate the antithrombogeneity of the HDVs in vitro.As we know, when a foreign surface comes into contactwith blood, the initial response is the adsorption of bloodproteins, followed by platelet adhesion and activation of the

coagulation pathways, leading to thrombus formation, there-fore, platelet adhesion is an important test for the evalua-tion of the blood compatibility of the grafts. Our results

FIGURE 5. Doppler ultrasound examination of the grafts at 3 months after implantation. (A) The occlusive control graft and (B) the patent TEVGs at 3

months after implantation. 22 � 7mm (300 � 300 DPI). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIGURE 6. 3-month patency rates of TEVGs and DVs grafts (both n ¼20). 69 � 54mm (600 � 600 DPI).

FIGURE 7. Histological analyses of intimal surface of the grafts

retrieved 3 months after implantation. (A) HE staining of midportion of

the retrieved TEVGs and (B) DVs graft (original magnification �50, white

arrows indicate endothelium, the scale bars indicate 200 lm). (C) SEM

of intimal surface of TEVGs and (D) DVs graft (original magnification

�500, the scale bars indicate 20 lm). (E) Masson trichrome staining

midportion of the retrieved TEVGs and (F) DVs graft (original magnifica-

tion �50, white arrows indicate neointimal formation, the scale bars

indicate 200 lm). 49 � 67mm (600 � 600 DPI). [Color figure can be

viewed in the online issue, which is available at wileyonlinelibrary.com.]



showed that platelets were rarely observed on the surfacesof HDVs, by contrast, the unheparinized DVs showed higherplatelet adhesion, which revealed HDVs possessed excellentability of anti-platelet adhesion. Moreover, previous resultsdemonstrated that HDVs eliminated blood clot formation invitro,4,26 and animal study indicated it reduced the throm-bogenicity in vivo as well.5

Establishing a reliable cell source is also vital for thesuccessful tissue engineering of vascular grafts. Recently,increased evidences have proved that EPCs have thepotential to differentiate into mature ECs and contributeto the process of endothelium repair.13,14 In present study,we harvested EPCs from canine PBMCs, after 3 weeks ofculture and proliferation, the induced cells presented atypical ‘‘spindle-shaped’’ structure, and the majority of thecells expressed the endothelial markers, von Willebrandfactor, Flk-1, and VE-cadherin. The cells were also capableof taking up LDL particles and UEA from the media simul-taneously. Additionally, by tube formation assays on aMatrigel matrix, we further confirmed the EPCs possessed

a high capacity to establish primitive vascular tube-likestructures.

In leading studies, EPCs have been seeded onto a varietyof vascular grafts, and primary good results were approvedin vivo experimentally.7,15 For example, Kaushal et al.7

reported that the EPCs-seeded vascular grafts remained pat-ent for 130 days as a carotid grafts in sheep, whereas non-seeded grafts occluded within 15 days, most importantly,the EPCs-seeded grafts exhibited contractile activity andnitric oxide mediated vascular relaxation, similar to nativecarotid arteries. He et al.15 seeded EPCs onto hybrid grafts(inner diameter of approximately 4.5 mm and length of 6cm), after 3 month of implantation, 11/12 engineered vas-cular grafts were patent. In agreement with those reports,our animal experiments demonstrated that TEVGs had a sig-nificantly higher patency rate than non-seeded grafts (19/20 vs. 12/20, p < 0.01) after 3 months of implantation. Allof the occluded grafts were found within first month afterimplantation, due to thrombus formation probably due tothe thrombogenicity of ECMs after decellularization. Wehypothesized that both EPCs seeding and heparin coatingcontributed to reduce the in vivo thrombogenicity, andimproved early patency rate of TEVGs. In addition, the his-tological analysis of the patent grafts retrieved at 3 monthsrevealed a high degree of reendothelialization in TEVGs and,conversely, only a few disorderly endothelium and eventhrombosis deposited on the luminal surface of the controlgrafts. Although we were not able to evaluate the reendo-thelialization at an earlier stage or analyze it quantitatively,the effects of EPCs on reendothelialization of vascular graftsin vivo were becoming increasingly clear,15 for example, Gri-ese et al observed significant endothelialization (40% to60% coverage) of the EPCs-seeded grafts compared with<5% in the unseeded grafts after 4 weeks implantation.27

The most important finding, in our opinion, was theobservation of the inhibition of neointimal hyperplasia inTEVGs. Figures 7 and 8 demonstrated a dramatic decreaseof intimal hyperplasia at both midportion and anastomoticsites of TEVGs compared with DVs (p < 0.001 for all sites).We hypothesized that the mechanism of neointimal hyper-plasia inhibition was principally through EPCs proliferating

FIGURE 8. Comparison of neointima area at midportion and anasto-

motic sites of TEVGs (n ¼ 19) and DVs (n ¼ 12) grafts at 3 months af-

ter implantation (* p < 0.0001 when compared to TEVGs at the same

site).80 � 61mm (300 � 300 DPI).

FIGURE 9. Immunofluorescent staining of the TEVGs retrieved 3 months after implantation. (A) CM-DiI-labeled cells (red) present in the intimal

sides and media parts of TEVGs (original magnification, �100). (B) vWF-positive cells (green) in the intima sides of TEVGs (original magnifica-

tion, �100). (C) Merged images of CM-DiI-labeled cells and vWF-positive cells (original magnification, �100). (All the scalebars indicate 100 um.)

39 � 13mm (300 � 300 DPI). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]


action on ECs which led to the reendothelialization of grafts,and that, this reendothelialization secondarily inhibited neo-intimal hyperplasia. As we know, ECs are able to secretemany antiproliferative factors, such as nitric oxide and vas-cular natriuretic peptides, which modulate the growth ofvascular smooth muscle cells (VSMCs). A lack of ECs mightpromote abnormal vascular growth, such as thrombosis,hyperplasia and obstruction. On the contrary, the presenceof normal ECs might lead to the normal growth of VSMCsand normal vascular regeneration in the long run. Moreover,we cannot exclude the effects of the immobilized heparin,which was found to inhibit the proliferation of VSMCs bothin vivo28 and in vitro,29 especially Cai et al.30 proved thatheparin coating delivered not only the antithrombogeneitybut also the antiproliferative property for decellularizedxenografts. Meanwhile, our findings were in substantialagreement with previous research. For instance, Griese et alshowed that EPCs transplantation led to early and nearlycomplete reendothelialization of the denuded carotid artery,resulting in inhibition of neointimal hyperplasia.27 However,the effect of EPCs on hyperplasical inhibination of TEVGswas for the first time reported in this study, it further con-firmed the feasibility of application of EPCs in constructingTEVGs. The present study also showed that CM-DiI-labeledcells were detected in the luminal sides and the medialparts of the explanted TEVGs after 3 months, which wasconsistent with the finding of previous study reporting thatcells labeled with CM-DiI can be observed up to 130 daysin vivo.7 The merged image of vWF and CM-DiI double-posi-tive cells suggested survival of the seeding EPCs and theircontribution to endothelium formation.

In conclusion, seeding EPCs onto HDVs might be anattractive approach for tissue engineering small-diametervascular grafts, and this technique seems to be able to gen-erate essentially normal vessel wall within a few months.

However, questions remain regarding how seeded EPCsproliferate, differentiate, and arrange themselves in anappropriate fashion to constitute a new tissue; further, ques-tions remain regarding decellularized scaffold degenerationand remodeling, suggesting additional animal studies will beneeded. In particular, experiments involving TEVG implanta-tion into growing animals would be necessary to determinewhether TEVGs have growth potential. Collectively, answersto these questions will contribute toward development offurther improved TEVGs.

REFERENCES

1. Zhou M, Liu Z, Wei ZQ, Liu C, Qiao T, Ran F, Bai Y, Jiang X, Ding

Y. Development and validation of a small diameter vascular tissue

from a decellularized scaffold coated with heparin and vascular

endothelial growth factor. Artif Organs 2009;33:230–239.

2. Roy S, Silacci P, Ste rgiopulos N. Biomechanical properties of

decellularized porcine common carotid arteries. Am J Physiol

Heart Circ Physiol 2005;289:H1567–H1576.

3. Gui L, Muto A, Chan SA, Breuer CK, Niklason LE. Development of

decellularized human umbilical arteries as small-diameter vascu-

lar grafts. Tissue Eng Part A 2009;15:2665–2676.

4. Conklin BS, Richter ER, Kreutziger KL, Zhong DS, Chen C. Devel-

opment and evaluation of a novel decellularized vascular xeno-

graft. Med Eng Phys 2002;24:173–183.

5. Wang XN, Chen CZ, Yang M, Gu YJ. Implantation of decellular-

ized small-diameter vascular xenografts with and without surface

heparin treatment. Artif Organs 2007;31:99–104.

6. Teebken OE, Bader A, Steinhoff G, Haverich A. Tissue engineering

of vascular grafts: human cell seeding of decellularised porcine

matrix. Eur J Vasc Endovasc Surg 2000;19:381–386.

7. Kaushal S, Amiel GE, Guleserian KJ, Shapira OM, Perry T, Suther-

land FW, Rabkin E, Moran AM, Schoen FJ, Atala A, Soker S, Bis-

choff J, Mayer JE Jr. Functional small diameter neovessels

created using endothelial progenitor cells expanded ex vivo. Nat

Med 2001;7:1035–1040.

8. Cho SW, Lim SH, Kim IK, Hong YS, Kim SS, Yoo KJ, Park HY,

Jang Y, Cahng BC, Choi CY, Hwang KC, Kim BS. Small-diameter

blood vessels engineered with bone marrow-derived cells. Ann

Surg 2005;241:506–515.

9. Zhao Y, Zhang S, Zhou J, Wang J, Zhen M, Liu Y, Chen J, Qi Z.

The development of a tissue-engineered artery using decellular-

ized scaffold and autologous ovine mesenchymal stem cells. Bio-

materials 2010;31:296–307.

10. Laube HR, Duwe J, Rutsch W, Konertz W. Clinical experience with

autologous endothelial cell-seeded polytetrafluoroethylene coronary

artery bypass grafts. J Thorac Cardiovasc Surg 2000;120:134–141.

11. Deutsch M, Meinhart J, Fischlein T, Preiss P, Zilla P. Clinical auto-

logous in vitro endothelialization of infrainguinal ePTFE grafts in

100 patients: A 9-year experience. Surgery 1999;126:847–855.

12. Shirota T, He H, Yasui H, Matsuda T. Human endothelial progeni-

tor cell-seeded hybrid graft: proliferative and antithrombogenic

potentials in vitro and fabrication processing. Tissue Eng 2003;9:

127–136.

13. Asahara T, Murohara T, Sullivan A, Silver M, van der Zee R, Li T,

Witzenbichler B, Schatteman G, Isner JM. Isolation of putative pro-

genitor endothelial cells for angiogenesis. Science 1997;275:964–967.

14. Shi Q, Rafii S, Wu MH, Wijelath ES, Yu C, Ishida A, Fujita Y,

Kothari S, Mohle R, Sauvage LR, Moore MA, Storb RF, Hammond

WP. Evidence for circulating bone marrow-derived endothelial

cells. Blood 1998;92:362–367.

15. He H, Shirota T, Yasui H, Matsuda T. Canine endothelial progeni-

tor cell-lined hybrid vascular graft with nonthrombogenic poten-

tial. J Thorac Cardiovasc Surg 2003;126:455–464.

16. Yao C, Roderfeld M, Rath T, Roeb E, Bernhagen J, SteffensG. The

impact of proteinase-induced matrix degradation on the release

of VEGF from heparinized collagen matrices. Biomaterials 2006;

27:1608–1616.

17. Spiller D, Losi P, Briganti E, Sbrana S, Kull S, Martinelli I, Soldani

G. PDMS content affects in vitro hemocompatibility of synthetic

vascular grafts. J Mater Sci Mater Med 2007;18:1097–1104.

18. Wu HK, Riha MG, Yang H, Li M, Yao Q, Chen C. Differentiation

and proliferation of endothelial progenitor cells from canine pe-

ripheral blood mononuclear cells. J Surg Res 2005;126:193–198.

19. Fuchs S, Hermanns MI, Kirkpatrick CJ. Retention of a differenti-

ated endothelial phenotype by outgrowth endothelial cells iso-

lated from human peripheral blood and expanded in long-term

cultures. Cell Tissue Res 2006;326:79–92.

20. Yang Z, Tao J, Wang JM, et al. Shear stress contributes to t-PA

mRNA expression in human endothelial progenitor cells and non-

thrombogenic potential of small diameter artificial vessels. Bio-

chem Biophys Res Commun 2006;342:577–584.

21. Sales KM, Salacinski HJ, Alobaid N, Mikhail M, Balakrishnan V,

Seifalian AM. Advancing vascular tissue engineering: the role of

stem cell technology. Trends Biotechnol 2005;23:461–467.

22. Heidenhain C, Weichert W, Schmidmaier G, Wildemann B, Hein

M, Neuhaus P, Heise M. Polymer coating of porcine decellularized

and cross-linked aortic grafts. J Biomed Mater Res B Appl Bio-

mater 2010;94:256–263.

23. Williams C, Liao J, Joyce EM, Wang B, Leach JB, Sacks MS,

Wong JY. Altered structural and mechanical properties in decellu-

larized rabbit carotid arteries. Acta Biomater 2009;5:993–1005.

24. Lehr EJ, Rayat GR, Chiu B, Churchill T, McGann LE, Coe JY, Ross DB.

Decellularization reduces immunogenicity of sheep pulmonary artery

vascular patches. J Thorac Cardiovasc Surg 2011;141:1056–1062.

25. Madden RL, Lipkowitz GS, Browne BJ, Kurbanov A. Experience

with cryopreserved cadaveric femoral vein allografts used for

hemodialysis access. Ann Vasc Surg 2004;18:453–458.



26. Liao D, Wang X, Lin PH, Yao Q, Chen CJ. Covalent linkage of hep-

arin provides a stable anti-coagulation surface of decellularized

porcine arteries. J Cell Mol Med 13:2736–2743, 2009.

27. Griese DP, Ehsan A, Melo LG, Kong D, Zhang L, Mann MJ, Pratt RE,

Mulligan RC, Dzau VJ. Isolation and transplantation of autologous

circulating endothelial cells into denuded vessels and prosthetic

grafts: Implications for cell-based vascular therapy. Circulation 2003;

108:2710–2715.

28. Clowes AW, Karnowsky MJ. Suppression by heparin of smooth

muscle cell-proliferation in injured arteries. Nature 1977;265:625–626.

29. Hoover RL, Rosenberg R, Haering W, Karnovsky MJ. Inhibition of

rat arterial smooth muscle cell proliferation by heparin. II. In vitro

studies. Circ Res 1980;47:578–583.

30. Cai WW, Gu YJ, Wang XN, Chen CZ. Heparin coating of small-cal-

iber decellularized xenografts reduces macrophage infiltration

and intimal hyperplasia. Artif Organs 2009;33:448–455.


tissue engineering of small-diameter vascular grafts by endothelial progenitor cells seeding...

Documents