Home
Search
Collections
Journals
About
Contact us
My IOPscience
In vivo endothelization of tubular vascular grafts through in situ recruitment of endothelial and endothelial progenitor cells by RGD-fused mussel adhesive proteins
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2015 Biofabrication 7 015007 (http://iopscience.iop.org/1758-5090/7/1/015007) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 193.255.248.150 This content was downloaded on 11/02/2015 at 14:39
Please note that terms and conditions apply.
Biofabrication 7 (2015) 015007
doi:10.1088/1758-5090/7/1/015007
PAPER
RECEIVED
2 June 2014 REVISED
1 December 2014
In vivo endothelization of tubular vascular grafts through in situ recruitment of endothelial and endothelial progenitor cells by RGDfused mussel adhesive proteins
ACCEPTED FOR PUBLICATION
9 December 2014 PUBLISHED
20 January 2015
Tae-Yun Kang1,5, Jung Ho Lee3,5, Bum Jin Kim4,5, Jo-A Kang3, Jung Min Hong2, Byoung Soo Kim2, Hyung Joon Cha4, Jong-Won Rhie3 and Dong-Woo Cho2 1 2 3 4
5
Department of Biomedical Engineering, Yale University, New Haven, CT 06520, USA Department of Mechanical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, 790-784, Korea Department of Plastic Surgery, College of Medicine, The Catholic University of Korea, Seoul, 137-701, Korea Department of Chemical Engineering, Pohang University of Science and Technology (POSTECH), San 31, Hyoja dong, Nam-gu, Pohang, Gyungbuk, 790-784, Korea Authors contributed equally to this study.
E-mail: [email protected], [email protected] and [email protected] Keywords: tissue engineering, vascular graft, endothelization, thrombosis Supplementary material for this article is available online
Abstract The use of tissue mimics in vivo, including patterned vascular networks, is expected to facilitate the regeneration of functional tissues and organs with large volumes. Maintaining patency of channels in contact with blood is an important issue in the development of a functional vascular network. Endothelium is the only known completely non-thrombogenic material; however, results from treatments to induce endothelialization are inconclusive. The present study was designed to evaluate the clinical applicability of in situ recruitment of endothelial cells/endothelial progenitor cells (EC/EPC) and preendothelization using a recombinant mussel adhesive protein fused with arginine–glycine–aspartic acid peptide (MAP-RGD) coating in a model of vascular graft implantation. Microporous polycaprolactone (PCL) scaffolds were fabricated with salt leaching methods and their surfaces were modified with collagen and MAP-RGD. We then evaluated their anti-thrombogenicity with an in vitro hemocompatibility assessment and a 4-week implantation in the rabbit carotid artery. We observed that MAP-RGD coating reduced the possibility of early in vivo graft failure and enhanced re-endothelization by in situ recruitment of EC/EPC (patency rate: 2/3), while endothelization prior to implantation aggravated the formation of thrombosis and/or IH (patency rate: 0/3). The results demonstrated that in situ recruitment of EC/EPC by MAP-RGD could be a promising strategy for vascular applications. In addition, it rules out several issues associated with pre-endothelization, such as cell source, purity, functional modulation and contamination. Further evaluation of long term performance and angiogenesis from the luminal surface may lead to the clinical use of MAP-RGD for tubular vascular grafts and regeneration of large-volume tissues with functional vascular networks.
Introduction Blood vasculature is viewed as a dynamic network that supplies tissues with nutrients and gases and regulates various processes through direct communication of endothelial cells (EC) with adjacent cells [1–3]. Recent advances in fabrication technology have enabled the formation of tissue mimics, including patterned vascular networks, by lining ECs in preconstructed © 2015 IOP Publishing Ltd
fluidic channels [4–8]. Studies of tissue mimics in vitro provide a better understanding of angiogenesis [5, 6, 9–12], cancer cell invasion/extravasation [9–11], and interactions between ECs and parenchymal cells [5, 9, 12]. Moreover, the use of the tissue mimics in vivo is expected to facilitate the regeneration of functional tissues and organs with large volumes, because surgical connection of the artificial vascular network to the host vasculature is the only tangible
Biofabrication 7 (2015) 015007
T-Y Kang et al
solution that will meet the metabolic needs of engineered tissue/organ constructs [13–15]. Therefore, maintaining patency of channels in contact with blood is important in developing a functional vascular network. Endothelium is the only known completely non-thrombogenic material, and previous studies have showed that the luminal endothelium prevents the activation of inflammatory and coagulatory events [16–19]. However, research in the field of small-diameter graft development found that successful seeding of EC does not necessarily result in a positive outcome clinically. Two main strategies are being explored to create endothelialized vascular grafts. The first is to seed cells in the graft and grow a confluent endothelial layer on the luminal surface before implantation. This technique has been studied with ECs from autologous veins [20] and adipose tissue [21, 22], or with endothelial progenitor cells (EPC) from bone marrow [23] and peripheral blood [24]. The second strategy is to induce endothelization in vivo by recruiting circulating EPCs through surface modifications with anti-CD34 antibodies [25] and RGD [26, 27]. Although some treatments enhanced endothelization and prevented early graft failure by thrombosis, they stimulated the formation of intimal hyperplasia (IH), especially at anastomotic sites, and induced EC and smooth muscle cell (SMC) dysfunction [22, 28, 29]. Our previous experience in vitro proved that the recombinant mussel adhesive protein fused with arginine–glycine–aspartic acid peptide (MAP-RGD) coating improved endothelization and inhibited platelet activation [19]. As the MAP-RGD coating can be easily applied to complex channels, the results suggested the potential of MAP-RGD for the development of a functional vascular network with channel patency. Therefore, we evaluated the clinical applicability of in situ recruitment of EC/EPC and pre-endothelization using a MAP-RGD coating in a model of vascular graft implantation.
4 mm by 8 mm pieces. To fabricate a tubular scaffold as a vascular graft, a 15-gauge needle was prepared as a mold; the outer diameter of this needle, 1.829 mm, corresponds to the inner diameter of the scaffold. The mold was dipped in the PCL mixture, placed in ambient air to extract chloroform, and then placed in deionized water to dissolve the sodium chloride particles. Finally, a tubular scaffold was prepared by taking off the needle and trimming the ends. The scaffold’s surfaces were modified with collagen and MAP-RGD as described previously [19]. Acid-soluble type I collagen (Koken atelocollagen implant, Koken, Japan) was dissolved in 0.1 M hydrochloric acid to a concentration of 1% (w/v). Scaffolds were immersed in collagen solution for 60 min at 4 °C for immobilization. The collagen-coated scaffolds were then crosslinked by using N-hydroxysuccinimide(NHS)/N-(3dimethylaminopropyl)-N’-ethylcarbodiimide (EDC) solution in ethanol (EDC/NHS at 1:1 each 10 mg ml−1 in 95% ethanol) for 3 h at 4 °C [34]. The remnant collagen was washed with distilled water three time. After crosslinking, the samples were freeze-dried and the process was repeated twice. For MAP immobilization, MAP-RGD solution was prepared by dissolving DOPA-modified fp-151-RGD in distilled water (DW) at 2 mg ml−1. Collagen-coated scaffolds were immersed in MAP-RGD solution for 24 h at room temperature then washed with phosphate buffered saline (PBS). A detailed description of MAP-RGD production and characteristics were previously reported [30–32]. For MAP immobilization, MAPRGD solution was prepared by dissolving 3,4-dihydroxyphenylalanine-modified fp-151-RGD in distilled water. We planned to adopt a collagen layer as a basement membrane on the luminal surface of an artificial vascular network in a porous scaffold. The collagen layer should be a barrier to maintain endothelial cells on the surface. Thus, the vascular grafts used in this study were also coated with collagen prior to MAP-RGD to provide the same environment as the artificial vascular network.
Materials and methods Scaffold fabrication and surface modification Thin films and tubular scaffolds were fabricated to evaluate the blood compatibility of surfaces in vitro and patency in vivo, respectively. A polycaprolactone (PCL; Polyscience, Inc., MW 43 000–50 000) solution was prepared in chloroform at a concentration of 10% (w/v). Milled sodium chloride particles (SigmaAldrich, USA) were sieved through a 45 μm mesh. These were then mixed with the PCL solution at a composition of 400 wt.% salt of PCL. To fabricate a porous and thin film, the PCL mixture was poured on a glass plate and dried in ambient air for 3 h. After dissolving the sodium chloride particles in water for 1 h, the film was peeled from the glass and cut into 2
Evaluation of mechanical properties Uniaxial tensile strength test was carried out to assess the effect of the amount of sodium chloride on graft mechanical properties using a test equipment (Instron, Norwood, MA, USA). The thin film specimens (25 mm × 12.5 mm × 300 μm) were prepared as described above and the sodium chloride particles were mixed with the PCL solution at a composition of 0, 100, 200, 400, and 600 wt.% salt of PCL. In the tensile test, the specimens were pulled at a crosshead velocity of 5 mm min−1. The mechanical tensile modulus was calculated from the stress-strain curves. In the stress-strain curve, the slope of the first linear portion was defined as the modulus of each specimen.
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 1. Perfusion system. Here, we show the configuration of the perfusion system (A), the perfusion chamber for the hemocompatibility test (B), and culture using tubular scaffolds (C).
In vitro hemocompatibility test with whole blood We employed a perfusion system (figure 1(A)) in an incubator with a 5% CO2 atmosphere at 37 °C to elucidate the interaction of blood and prepared samples. The system configuration was described in our previous study [19]. Briefly, the system was composed of a peristaltic pump, reservoir, gas exchanger, and a perfusion chamber designed to load thin films (figure 1(B)). Fresh blood was collected with the use of Vacutainer tubes containing sodium citrate (BD, USA) from a healthy adult volunteer free of aspirin or other drugs that could bias the results. Prior to blood circulation, all components of the perfusion system and samples were sterilized for 2 h with 70% ethanol and ultraviolet (UV) light. Any remaining ethanol was removed by washing with PBS. The flow rate was set to match physiological shear stress levels [33] (1 Pa) in a carotid artery of a rabbit. Whole blood was circulated for 3 h in the perfusion system, after which the scaffolds, in the form of thin films, were taken out and fixed in 10% formalin for 30 min. During the whole blood circulation for 3 h, blood wasn’t coagulated in the system. However, in our experience, it started to be coagulated after 8 h. Immunohistochemical staining was used to visualize the adhered platelets, leukocytes and EPCs using antibodies to CD41 (abcam, USA), CD45 (abcam, USA), and CD34 (biorbyt Ltd, UK), respectively. All images were obtained using a confocal microscope (Olympus confocal FluoView 1000). And the coverage ratio of platelets and the numbers of leukocytes/EPCs were quantified by imageJ program with3 images acquired from different sites. In an image, cells stained 3
with DAPI/ CD45 were counted as leukocyte but cells stained with DAPI/CD 34 were counted as EPC. Endothelial cell isolation and characterization ECs were isolated by following a protocol described by Russell et al [34]. A male rabbit (3.0 kg) was sacrificed by isoflurane inhalation. The thoracic aorta was excised immediately from the heart after sacrifice and rinsed in PBS. The aorta was transferred to a dish filled with a collagenase solution and incubated for 5–10 min at 37 °C. The endothelial layer was then gently removed by rotating a dry, sterile swab onto the surface of the aorta and dabbing the swab in the collagenase solution to dislodge cells from the tip. Cells were collected by centrifugation at 1000 rpm for 5 min and the supernatant was aspirated. After adding 3 mL of endothelial growth medium (EGM-2) with 10% fetal bovine serum (Thermo Scientific HyClone, Waltham, MA) and 1% antibiotic/antimycotic solution (Gibco, Carlsbad, CA), the cell suspension was centrifuged again. The cells were cultured in the same medium in an incubator with a 5% CO2 atmosphere at 37 °C, and used at passage 4 for all experiments. Immunohistochemical staining was used to identify the isolated cells with antibodies to von Willebrand factor (vWf, biorbyt Ltd, UK) and α-smooth muscle actin (α-SMA, abcam, USA). All images were obtained using a confocal microscope. Cell seeding and perfusion culture Segments of a silicone tube were prepared as a perfusion chamber for tubular scaffolds (figure 1(C)). After expansion, an EC suspension was injected into
Biofabrication 7 (2015) 015007
T-Y Kang et al
the perfusion chamber with fluorescent labeling (Neostem, South Korea), followed by shaking and incubating for 3 h. The chamber was then connected to the perfusion system in an incubator with a 5% CO2 atmosphere at 37 °C. Cell-seeded scaffolds were cultured under perfusion conditions for 6 days, while the shear stress was increased from 0.1 Pa to 1 Pa to maximize the proliferation rate, as shown in our previous study [19]. In vivo implantation and sacrifice The experimental protocol was approved by the Catholic University of Korea Animal Care and Use Committee. Fifteen New Zealand White rabbits (2.5–3.0 kg) were divided into 5 groups: bare PCL, bare PCL/Collagen, bare PCL/Collagen/MAP-RGD, cell-seeded PCL/Collagen, and cell-seeded PCl/Collagen/MAP-RGD. Under general anesthesia by intravenous injection of combination of intraperitoneal ketamine (75 mg kg−1) and xylazine (10 mg kg −1), the neck skin was shaved and the right carotid artery was exposed. The artery was clamped at the proximal and distal ends and then transected. The tubular scaffold was trimmed to 8mm in length and sutured end to end with 8 to 10 interrupted stiches using a 9-0 ethilon suture. No anticoagulation or anti-platelet drugs were given before or after the implantation procedure. After 4 weeks, the animals were reanesthetized and the patency of scaffolds was observed by restoration of blood flow in the distal vessel after squeezing and releasing the vessel with forceps. More specifically, one pair of forceps was used to hold the distal blood vessel near the anastomotic site, while the other pair of forceps was used to squeeze the vessel toward the distal end, so the vessel became flat. We then released the forceps near the anastomotic sites to observe that blood flow could be restored, inflating the vessel. Scaffolds were subsequently resected and washed with heparinized saline to remove the remaining blood.
Histological analysis Grafts were explanted after 4 weeks, fixed in 10% formalin for 30 min, and divided into proximal, middle, and distal regions. The pieces were separately embedded in O.C.T. compound (Tissue-Tek, Sakura Fineteck Inc., Torrance, CA, USA) and 4 μm cryosections were taken for histological evaluation. The proximal and distal regions were sectioned longitudinally, but the middle regions were sectioned transversely. All sections were removed from O.C.T. using 10% formalin and stained with hematoxylin and eosin (H&E). Immunohistochemical staining was used to analyze the sections with primary antibodies to vWf (biorbyt, UK) and α-SMA (abcam, USA). All images were obtained using a confocal microscope. 4
1.1. Statistical analysis Statistical analyses were performed using one-way analysis of variance (ANOVA) with a post-hoc Tukey test using MINITAB (version 14.2). Differences between groups were considered statistically significant at P < 0.05.
Results Microporous thin films and tubular scaffolds Thin films (figure 2(A)) and tubular scaffolds (figures 2(B) and (C)) were successfully fabricated using salt-leaching methods. As shown in figure 2(D), the scaffolds exhibited irregular pores on the surface. After collagen and MAP-RGD coating, the pores were covered by a thin layer of collagen (figure 2(E)) and the flat surface became grainy in sequence (figure 2(F)). Uniaxial tensile modulus according to the amount of sodium chloride Figure 2(G) shows the tensile modulus changes according to the amount of sodium chloride. The tensile strength of PCL specimen was about 235 MPa, which is much higher than those of natural vessels. However, microporous specimens fabricated by salt leaching method showed significantly decreased values. When the salt/polymer ratio was 4 (400%) to 6 (600%), the tensile modulus values were comparable to the values of the natural vessels [35–37]. By considering the ease of handling for the microsurgical anastomosis, 400 wt.% salt of PCL was used for the in vivo implantation. Hemocompatibility of bare PCL, PCL/Collagen, and PCL/Collagen/MAP-RGD The contact of blood with biomaterials triggers the coagulation and complement cascades, resulting in thrombosis and inflammation [38]. The ideal biomaterial for vascular applications exhibits low thrombogenicity and rapid re-endothelization when in contact with circulating blood. Therefore, we characterized the hemocompatibility of the samples based on the adhesion of platelets (figures 3(A)–(C)), leukocytes (figures 3(D)–(F)) and EPCs (figures 3(G)-(I)). As shown in figure 3(J), there was no statistically significant difference in the number of adhered platelets on the PCL, PCL/Collagen, and PCL/Collagen/MAPRGD grafts. However, PCL/Collagen/MAP-RGD grafts had fewer leukocytes (figure 3(K)), but markedly more EPCs (figure 3(L)) on the surface than PCL and PCL/Collagen grafts. Characteristics of isolated endothelial cells Figure 4(A) shows the morphology of the cells isolated form a heart artery and passaged up to four times. Some of the cells grew in a ‘cobblestone morphology’, which is characteristic of EC. However, some grew in a
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 2. Structural characterization of thin films and grafts. Optical image of fabricated PCL thin film (A) and tubular grafts (B); scanning electron microscopy image for the cross-section the graft (C); luminal surface of PCL (D), PCL/Collagen (E), and PCL/ Collagen/MAP-RGD grafts (F); (G) tensile modulus of thin films according to the amount of sodium chloride.
‘spindle-shaped’ pattern, which is the characteristic of SMC. To confirm that the isolated cells were contaminated by SMCs, double-staining was carried out with vWf (biorbyt Ltd, UK) and α-SMA (abcam, USA), which are specific markers for EC and SMC, respectively. Figure 4(B) indicates that SMCs were still present with ECs even at passage 4. In vivo patency of implanted grafts To compare the anti-thrombogenic properties of vascular grafts in vivo, we implanted bare PCL, bare PCL/Collagen, bare PCL/Collagen/MAP-RGD grafts, cell-seeded PCL/Collagen, and cell-seeded PCL/Collagen/MAP-RGD grafts into the common carotid artery of rabbits for 4 weeks. Patency was determined by restoration of blood flow in the distal vessel after squeezing and releasing the vessel with forceps (Movie S1-S2) and histological analysis (figures 5(C)–(F)). Among bare grafts, 3 of 3 PCL and 2 of 3 PCL/ Collagen/MAP-RGD grafts were patent. As one of the PCL/Collagen/MAP-RGD grafts was broken, the 5
patency could not be observed. The failed PCL/ Collagen/MAP-RGD graft was found to be broken into pieces and surrounded by fibrosis. All bare PCL/ Collagen, cell-seeded PCL/Collagen/MAP-RGD, cellseeded PCL/Collagen/MAP-RGD grafts showed no restoration of blood flow and were therefore considered to be completely occluded. H&E images of bare PCL and PCL/Collagen/MAP-RGD grafts show that the middle region and proximal/distal anastomotic sites remained patent with little neointimal formation. On the other hand, the occluded bare PCL/Collagen, cell-seeded PCL/Collagen, and cell-seeded PCL/Collagen/MAP-RGD grafts were clogged due to severe thrombosis and IH. In vivo endothelization and intimal hyperplasia Immunofluorescent staining revealed signs of IH and/ or endothelization on the luminal walls of the grafts (figures 5(A)–(F)). The patent grafts (bare PCL and bare PCL/Collagen/MAP-RGD) showed different results. The bare PCL grafts had evenly distributed IH
Biofabrication 7 (2015) 015007
T-Y Kang et al
Coverage ratio [%]
J
Cell Number
K
Cell Number
L
18 16 14 12 10 8 6 4 2 0
45 40 35 30 25 20 15 10 5 0
Platelet Adhesion
PCL
PCL/Collagen
PCL/Collagen/MAP-RGD
Leukocyte adhesion
PCL
PCL/Collagen PCL/Collagen/MAP-RGD
EPC adhesion
60 50 40 30 20 10 0
PCL
PCL/Collagen PCL/Collagen/MAP-RGD
Figure 3. Blood-material interaction in vitro. Platelets adhered to PCL (A), PCL/Collagen (B), and PCL/Collagen/MAP-RGD (C) are in green, labeled with CD41. Leukocytes adhered on PCL (D), PCL/Collagen (E), PCL/Collagen/MAP-RGD (F) are in red, labeled with CD45. EPCs adhered on PCL (G), PCL/Collagen (H), PCL/Collagen/MAP-RGD (I) are in green, labeled with CD34. We quantified adhered platelets (J), leukocytes (K), and EPCs (L) from the images.
Figure 4. Characterization of isolated ECs showed the microscopic appearance of isolated cells at passage 4 (A), ECs staining positive for the von Willebrand factor, and smooth muscle cells staining positive for α-smooth muscle actin (B).
that was stained with α-SMA, but we could not observe regenerated endothelium in the lumen (figure 6(A)). On the other hand, the bare PCL/Collagen/MAPRGD had non-uniform IH, which was rarely present in one region, but thicker than that of bare PCL grafts in another region. Interestingly, we observed endothelial regeneration in the lumen even though endothelial coverage was not complete (figure 6(C)). The occluded grafts (bare PCL/Collagen, cell-seeded PCL/ Collagen, and cell-seeded PCL/Collagen/MAP-RGD) were filled with endothelial cells and smooth muscle cells. Here, we present an image of a bare PCL/ Collagen graft (figure 6(B)) as a representative occluded graft. Before implantation, the cell-seeded grafts were completely covered with primary 6
endothelial cells, as shown in figure 6(D). However, the fluorescently (Neo-stem, South Korea) labeled cells were present in the occluded region of the cellseeded PCL/Collagen/MAP-RGD grafts (figure 6(F)), but not in cell-seeded PCL/Collagen grafts (figure 6(E)). The results indicated that the seeded cells were retained by MAP-RGD, but they actually aggravated the formation of thrombosis and/or IH.
Discussion The unique capacity of EC as an anti-thrombotic surface renders re-endothelization to be an effective approach for vascular applications. However, previous research has shown that vascular biology is more
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 5. An image taken immediately after implantation of a graft (A); illustration of graft regions for histological analysis (B); representative H&E staining images of bare PCL (C), bare PCL/Collagen (D), bare PCL/Collagen/MAP-RGD (E), cell-seeded PCL/ Collagen (F), and cell-seeded PCL/Collagen/MAP-RGD (G). The blue arrows indicate the implanted graft structures. The tissues formed inside of the grafts represent the intimal hyperplasia.
7
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 6. Characterization of the vascular neotissues by immunofluorescent staining. Formations of the smooth muscle layer and endothelial layer stained with α-SMA and vWf, respectively, on bare PCL (A), bare PCL/Collagen (B), and bare PCL/Collagen/MAPRGD (C) grafts, the thickness of SMC layer represents the extent of SMC invasion. The bare PCL/Collagen/MAP-RGD had nonuniform IH and a newly formed endothelial layer; cells covering the inner surface of a graft before implantation (D); absence of green labeled cells indicating the loss of in vitro seeded cells in a PCL/Collagen graft (E); presence of green labeled cells indicating the retention of in vitro seeded cells in a PCL/Collagen/MAP-RGD graft (F); white dotted lines indicate the interface of microporous PCL part and thrombosis/IH.
complex than originally believed, and it is still a challenge to enhance the formation of endothelium while preventing the activation of inflammatory/ thrombotic events and abnormal IH [22, 23, 28]. In this regard, surface modification with MAP-RGD is a valuable strategy to improve graft patency. In vitro hemocompatibility assessment of bare PCL, PCL/Collagen, and PCL/Collagen/MAP-RGD revealed several interesting points that influenced the in vivo vascular patency and organization. The interaction of bare PCL/Collagen/MAP-RGD with blood was compared with the results from bare PCL and bare PCL/Collagen. PCL is a well-known biocompatible polymer [39] and it is an adequate material for vascular applications because of its excellent mechanical properties and slow degradation rate in vivo. Although vascular grafts made of PCL showed a promising patency rate [39, 40], modifications of PCL grafts have been attempted to enhance the adhesion of vascular cells for long term performance [41–43]. Collagen, a natural extracellular matrix polymer, is favorable for cell adhesion, but it is a highly thrombogenic material [44]. Bare PCL/Collagen/MAP-RGD had better hematocompatibility than the other two groups. Bare PCL/Collagen/MAP-RGD also had fewer leukocytes but more EPCs on the surface than bare PCL or bare PCL/Collagen. Although there was no statistical difference in platelet adhesion, we previously found that 8
platelets were not activated on PCL/Collagen/MAPRGD [19]. Consistently, in vivo studies showed that a MAPRGD coating reduced the possibility of early in vivo graft failure and enhanced re-endothelization by in situ recruitment of EC/EPCs. The patency rate of bare PCL/Collagen/MAP-RGD grafts (2/3) was lower than that of bare PCL grafts (3/3) because the exceptionally broken PCL/Collagen/MAP-RGD graft was considered as a failure. However, histological staining showed that bare PCL/Collagen/MAP-RGD had more potential as a graft in vascular regeneration than bare PCL. During its implantation, the animal slightly awoke from the anesthesia and moved over. We assumed that the graft had been damaged at that moment and broken in the end. Although we followed a standard protocol of anesthesia, there was still an individual difference for each animal. Therefore, more delicate efforts will be required to adjust the anesthetic drugs. After the implantation of vascular grafts in vivo, the primary change seen in the lumen is IH, which reflects the accumulation of inflammatory, coagulatory, and hemodynamic factors [25, 45–47]. IH consists predominantly of smooth muscle cells, and excessive IH usually leads to graft failure [28, 48]. Bare PCL grafts showed spatially homogeneous IH at the middle region and proximal/distal anastomotic sites. However, bare PCL/Collagen/MAP-RGD grafts showed less IH than PCL at proximal and distal
Biofabrication 7 (2015) 015007
T-Y Kang et al
anastomotic sites, where IH usually develops significantly more than in the graft body [49]. As shown in figure2(D), MAP-RGD doesn’t cover the whole surface. In our previous research [19], on the whole, the flat surface had become grainy by MAP-RGD. Compared with the flat sample used in the previous research, more efforts is required to coat MAP-RGD on the surface of the tubular graft. The spatially nonhomogeneous IH formation at the middle region of the PCL/Collagen/MAP-RGD graft might be caused by the nonhomogeneous MAP-RGD coating. We expect that the IH formation at the middle region could be reduced by increasing the number of MAPRGD coatings prior to implantation. Moreover, MAPRGD has been proven to accelerate re-endothelization, which is a crucial factor in maintaining the long term patency of a graft. Progress of EC coverage was found even at the middle region, remote from anastomotic sites of bare PCL/Collagen/MAP-RGD graft, whereas there was no EC coverage at the same region of bare PCL grafts. And the extent of SMC invasion was inhibited by the newly formed endothelium on PCL/Collagen/MAP-RGD graft. Re-endothelization of a graft could occur via out-growth of local endothelial cells from the host vessels adjacent to anastomotic sites or via recruitment of circulating endothelial progenitor cells as the re-endothelization of injured arteries [50]. MAP-RGD coating could have positive influences on both of mechanisms of re-endothelization. Previous studies have reported the capacity of RGD to promote the attachment of ECs [27, 43], capture circulating EPCs [51], and inhibit platelet adhesion/activation [19, 43]. In addition, our present study showed that MAP-RGD reduced the adhesion of leukocytes in blood. Based on these results, the present in vitro and in vivo study demonstrated that the fusion of RGD with MAP did not hamper intrinsic hemocompatibility while improving the immobility of RGD. If MAP-RGD coating is applied to the construction of a functional vascular network, EC/EPC recruitment by MAP-RGD could accelerate angiogenesis not only from the luminal surface of a vascular network, but also from the outside of the scaffold. Here, we observed that endothelization prior to implantation aggravated the formation of thrombosis and/or IH. Using a fluorescent dye, it has been shown that after 4 weeks of follow-up, seeded cells remained in PCL/Collagen/MAP-RGD grafts due to the strong binding affinity of MAP-RGD, but not in PCL/Collagen. However, the retained cells were found in the occluded region and it is suggested that the cells contributed to thrombosis and/or IH. The failure of the cell-seeded graft could be caused by complicated factors occurring in any step of the procedures for harvest, expansion, and perfusion culture. Although we carefully swabbed the luminal surface of the vessel to harvest only ECs, a number of SMCs were present and we seeded cells without purification. Many researchers have tried to develop effective techniques to isolate 9
and purify endothelial cells. However, our study was designed to evaluate the feasibility of MAP-RGD for in situ recruitment of endothelial cells and compare it with risky environments by pre-seeded primary cells. Although it will take more time to purify and expand cells, delicately purified cells could increase the patency of vascular grafts. Second, the exposure of ECs to media containing fetal bovine serum possesses the potential for both genotypic and phenotypic modulation of EC during in vitro expansion and perfusion culture. In addition, the human manipulation of the perfusion culture system could introduce the risk of contamination. We actually went through contamination during the initial experiments. Perfusion system consists of several components and it has high risk of contamination in each step such as assembling the whole system, transferring it from a bench to incubation. At the initial stage, we made efforts to get rid of unnecessary steps and ensure the sterilized status. The present study was designed to evaluate the clinical applicability of MAP-RGD coating in the construction of a functional vascular network for large tissue regeneration. Therefore, the graft patency was observed for a relatively short term with the assumption that blood delivery by the vascular network to the interior could be replaced by angiogenesis in the scaffold during the period. However, we cannot guarantee the patency of the vascular network in the same period of this work, because a vascular network could have smaller and more complex geometry than a simple vascular graft. Moreover, long term graft performance such as patency, structural integrity, calcification, polymer degradation, and vascular remodeling needs to be evaluated to create clinically acceptable vascular prostheses. Especially, the mismatch of mechanical properties between an engineered graft and native artery could cause excessive intimal hyperplasia of the connected artery [52]. Although the data were not presented here, in our initial experiment, the native vessels, which were connected to relatively rigid tubular grafts, were found to be not only occluded but also hardened. Further investigation will be required to elucidate the effect of mechanical compliance on the graft patency and integrity to the native vessel.
Conclusion This study evaluated the clinical applicability of a MAP-RGD coating in a model of vascular graft implantation. Here, we observed that MAP-RGD coating reduced the possibility of early in vivo graft failure and enhanced re-endothelization by in situ recruitment of EC/EPC (patency rate: 2/3), while endothelization prior to implantation aggravated the formation of thrombosis and/or IH (patency rate: 0/ 3). The results demonstrated that in situ recruitment of EC/EPC by MAP-RGD could be a promising strategy for vascular application that rules out several
Biofabrication 7 (2015) 015007
T-Y Kang et al
issues of pre-endothelization such as cell source, purity, functional modulation, and contamination. Further evaluation of the long term performance and angiogenesis from the luminal surface may potentially lead to the clinical use of MAP-RGD not just for tubular vascular grafts, but for regeneration of large volume tissues with functional vasculature.
[16]
[17]
[18]
Acknowledgments This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (No. 2012-0001235 and No. 2011-0030075) (to D.-W. C), Marine Biotechnology Program funded by the Ministry of Oceans and Fisheries, Korea (to H.J.C) and the Rising Star Program funded by POSTECH (to H.J.C).
[19]
[20]
[21]
References [1] Cines D B et al 1998 Endothelial cells in physiology and in the pathophysiology of vascular disorders Blood 91 3527–61 [2] Cleaver O and Melton D A 2003 Endothelial signaling during development Nat. Med. 9 661–8 [3] Lazarus A, Del-Moral P M, Ilovich O, Mishani E, Warburton D and Keshet E 2011 A perfusion-independent role of blood vessels in determining branching stereotypy of lung airways Development 138 2359–68 [4] Choi N W, Cabodi M, Held B, Gleghorn J P, Bonassar L J and Stroock A D 2007 Microfluidic scaffolds for tissue engineering Nat. Mater. 6 908–15 [5] Miller J S et al 2012 Rapid casting of patterned vascular networks for perfusable engineered three-dimensional tissues Nat. Mater. 11 768–74 [6] Zheng Y et al 2012 In vitro microvessels for the study of angiogenesis and thrombosis Proc. Natl. Acad. Sci. USA 109 9342–7 [7] Jung J W, Kang H W, Kang T Y, Park J H, Park J and Cho D W 2012 Projection image-generation algorithm for fabrication of a complex structure using projection-based microstereolithography Int. J. Precis. Eng. Man. 13 445–9 [8] Seol Y J, Kang T Y and Cho D W 2012 Solid freeform fabrication technology applied to tissue engineering with various biomaterials Soft. Matter 8 1730–5 [9] Chung S, Sudo R, Vickerman V, Zervantonakis I K and Kamm R D 2010 Microfluidic platforms for studies of angiogenesis, cell migration, and cell-cell interactions Ann. Biomed. Eng. 38 1164–77 [10] Zervantonakis I K, Hughes-Alford S K, Charest J L, Condeelis J S, Gertler F B and Kamm R D 2012 Threedimensional microfluidic model for tumor cell intravasation and endothelial barrier function Proc. Natl. Acad. Sci. USA 109 13515–20 [11] Jeon J S, Zervantonakis I K, Chung S, Kamm R D and Charest J L 2013 In vitro model of tumor cell extravasation PLOS One 8 e56910 [12] Huh D, Matthews B D, Mammoto A, Montoya-Zavala M, Hsin H Y and Ingber D E 2010 Reconstituting organ-level lung functions on a chip Science 328 1662–8 [13] Baptista P M, Siddiqui M M, Lozier G, Rodriguez S R, Atala A and Soker S 2011 The use of whole organ decellularization for the generation of a vascularized liver organoid Hepatology 53 604–17 [14] Arenas-Herrera J E, Ko I K, Atala A and Yoo J J 2013 Decellularization for whole organ bioengineering Biomed. Mater. 8 014106 [15] Mirmalek-Sani S H, Deegan D B, Sullivan D C, Ko I, George S K, Jackson J D, Aboushwareb T A, Atala A and Yoo J
10
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29] [30]
[31]
[32]
[33]
[34] [35]
2012 Optimizing decellularization of porcine kidneys for whole organ bioengineering J. Tissue Eng. Regen. M 6 172 Blit P H, McClung W G, Brash J L, Woodhouse K A and Santerre J P 2011 Platelet inhibition and endothelial cell adhesion on elastin-like polypeptide surface modified materials Biomaterials 32 5790–800 Knetsch M L W, Aldenhoff Y B J, Schraven M and Koole L H 2004 Human endothelial cell attachment and proliferation on a novel vascular graft prototype J. Biomed. Mater. Res. A 71A 615–24 Wu H C, Wang T W, Kang P L, Tsuang Y H, Sun J S and Lin F H 2007 Coculture of endothelial and smooth muscle cells on a collagen membrane in the development of a smalldiameter vascular graft Biomaterials 28 1385–92 Kang T Y, Hong J M, Kim B J, Cha H J and Cho D W 2013 Enhanced endothelialization for developing artificial vascular networks with a natural vessel mimicking the luminal surface in scaffolds Acta Biomater. 9 4716–25 Lamm P, Juchem G, Milz S, Schuffenhauer M and Reichart B 2001 Autologous endothelialized vein allograft—a solution in the search for small-caliber grafts in coronary artery bypass graft operations Circulation 104 I108–14 Koyama M, Satoh K, Yoshida H, Suzuki S, Koie H and Takamatsu S 1996 Surface coverage of vascular grafts with cultured human endothelial cells from subcutaneous fat tissue obtained with a biopsy needle Thromb. Haemostasis. 76 610–4 Arts C H P, Joosten P P A H, Blankensteijn J D, Staal F J T, Ng P Y Y, Heijnen-Snyder G J, Sixma J J, Verhagen H J M, de Groot P G and Eikelboom B C 2002 Contaminants from the transplant contribute to intimal hyperplasia associated with microvascular endothelial cell seeding Eur. J. Vasc. Endovasc. 23 29–38 Jobst B J, Riegger G A J and Griese D P 2009 Endothelial cell seeding fails to prevent intimal hyperplasia following arterial injury in the rat carotid model Cardiovasc Drug Ther. 23 343–53 Brunt K R, Hall S, Ward C A and Melo L G 2007 Endothelial progenitor cell and mesenchymal stem cell isolation, characterization, viral transduction Methods Mol. Med. 139 197–210 Rotmans J I, Heyligers J M M, Verhagen H J M, Velema E, Nagtegaal M M, de Kleijn D P V, de Groot F G, Stroes E S G and Pasterkamp G 2005 In vivo cell seeding with anti-cd34 antibodies successfully accelerates endothelialization but stimulates intimal hyperplasia in porcine arteriovenous expanded polytetrafluoroethylene grafts Circulation 112 12–8 Li J H, Ding M M, Fu Q, Tan H, Xie X Y and Zhong Y P 2008 A novel strategy to graft rgd peptide on biomaterials surfaces for endothelization of small-diamater vascular grafts and tissue engineering blood vessel J. Mater. Sci-Mater. M 19 2595–603 Tugulu S, Silacci P, Stergiopulos N and Klok H A 2007 Rgd— functionalized polymer brushes as substrates for the integrin specific adhesion of human umbilical vein endothelial cells Biomaterials 28 2536–46 Subbotin V M 2007 Analysis of arterial intimal hyperplasia: review and hypothesis Theor. Biol. Med. Model 4 Loscalzo J 2000 Vascular matrix and vein graft failure—is the message in the medium? Circulation 101 221–3 Choi B H, Choi Y S, Kane D G, Kim B J, Song Y H and Cha H J 2010 Cell behavior on extracellular matrix mimic materials based on mussel adhesive protein fused with functional peptides Biomaterials 31 8980–8 Hwang D S, Gim Y, Yoo H J and Cha H J 2007 Practical recombinant hybrid mussel bloadhesive fp-151 Biomaterials 28 3560–8 Hwang D S, Sim S B and Cha H J 2007 Cell adhesion biomaterial based on mussel adhesive protein fused with rgd peptide Biomaterials 28 4039–46 Matsumoto T, Okumura E, Miura Y and Sato M 1999 Mechanical and dimensional adaptation of rabbit carotid artery cultured in vitro Med. Biol. Eng. Comput. 37 252–6 Gould R A and Butcher J T 2010 Isolation of valvular endothelial cells J. Visualized Exp. 46 e2158 Williams C, Liao J, Joyce E M, Wang B, Leach J B, Sacks M S and Wong J Y 2009 Altered structural and
Biofabrication 7 (2015) 015007
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
T-Y Kang et al
mechanical properties in decellularized rabbit carotid arteries Acta Biomater. 5 993–1005 Hashi C K, Derugin N, Janairo R R, Lee R, Schultz D, Lotz J and Li S 2010 Antithrombogenic modification of small-diameter microfibrous vascular grafts Arteriosclerosis, Thrombosis, And Vascular Biology 30 1621–7 Hasan A, Memic A, Annabi N, Hossain M, Paul A, Dokmeci M, Dehghani F and Khademhosseini A 2014 Electrospun scaffolds for tissue engineering of vascular grafts Acta Biomater. 10 11–25 Gorbet M B and Sefton M V 2004 Biomaterial-associated thrombosis: Roles of coagulation factors, complement, platelets and leukocytes Biomaterials 25 5681–703 de Valence S, Tille J C, Mugnai D, Mrowczynski W, Gurny R, Moller M and Walpoth B H 2012 Long term performance of polycaprolactone vascular grafts in a rat abdominal aorta replacement model Biomaterials 33 38–47 Mathews A, Colombus S, Krishnan V K and Krishnan L K 2012 Vascular tissue construction on poly(e-caprolactone) scaffolds by dynamic endothelial cell seeding: effect of pore size J. Tissue Eng. Regen. M 6 451–61 Zhang M, Wang K, Wang Z X, Xing B, Zhao Q and Kong D L 2012 Small-diameter tissue engineered vascular graft made of electrospun pcl/lecithin blend J. Mater. Sci-Mater. M 23 2639–48 Wissink M J B, Beernink R, Poot A A, Engbers G H M, Beugeling T, van Aken W G and Feijen J 2000 Improved endothelialization of vascular grafts by local release of growth factor from heparinized collagen matrices J. Control Release 64 103–14 Zheng W T et al 2012 Endothelialization and patency of rgdfunctionalized vascular grafts in a rabbit carotid artery model Biomaterials 33 2880–91
11
[44] Cloft H J et al 2000 Bovine type i collagen as an endovascular stent-graft material: biocompatibility study in rabbits Radiology 214 557–62 [45] Mattana J, Effiong C, Kapasi A and Singhal P C 1997 Leukocyte-polytetrafluoroethylene interaction enhances proliferation of vascular smooth muscle cells via tumor necrosis factor-alpha secretion Kidney Int. 52 1478–85 [46] Ojha M 1997 In vitro studies on anastomotic intimal hyperplasia Asaio J. 43 242 [47] Roy-Chaudhury P, Kelly B S, Miller M A, Reaves A, Armstrong J, Nanayakkara N and Heffelfinger S C 2001 Venous neointimal hyperplasia in polytetrafluoroethylene dialysis grafts Kidney Int. 59 2325–34 [48] Desai M, Mirzay-Razzaz J, von Delft D, Sarkar S, Hamilton G and Seifalian A M 2010 Inhibition of neointimal formation and hyperplasia in vein grafts by external stent/ sheath Vasc Med. 15 287–97 [49] Ojha M, Leask R L, Johnston K W, David T E and Butany J 2000 Histology and morphology of 59 internal thoracic artery grafts and their distal anastomoses Ann. Thorac. Surg. 70 1338–44 [50] Van der Heiden K, Gijsen F J, Narracott A, Hsiao S, Halliday I, Gunn J, Wentzel J J and Evans P C 2013 The effects of stenting on shear stress: relevance to endothelial injury and repair Cardiovasc Res. 99 269–75 [51] Kim K L et al 2009 Enhanced dermal wound neovascularization by targeted delivery of endothelial progenitor cells using an rgd-g-plla scaffold Biomaterials 30 3742–8 [52] Seifu D G, Purnama A, Mequanint K and Mantovani D 2013 Small-diameter vascular tissue engineering Nat. Rev. Cardiology 10 410–21
Search
Collections
Journals
About
Contact us
My IOPscience
In vivo endothelization of tubular vascular grafts through in situ recruitment of endothelial and endothelial progenitor cells by RGD-fused mussel adhesive proteins
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2015 Biofabrication 7 015007 (http://iopscience.iop.org/1758-5090/7/1/015007) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 193.255.248.150 This content was downloaded on 11/02/2015 at 14:39
Please note that terms and conditions apply.
Biofabrication 7 (2015) 015007
doi:10.1088/1758-5090/7/1/015007
PAPER
RECEIVED
2 June 2014 REVISED
1 December 2014
In vivo endothelization of tubular vascular grafts through in situ recruitment of endothelial and endothelial progenitor cells by RGDfused mussel adhesive proteins
ACCEPTED FOR PUBLICATION
9 December 2014 PUBLISHED
20 January 2015
Tae-Yun Kang1,5, Jung Ho Lee3,5, Bum Jin Kim4,5, Jo-A Kang3, Jung Min Hong2, Byoung Soo Kim2, Hyung Joon Cha4, Jong-Won Rhie3 and Dong-Woo Cho2 1 2 3 4
5
Department of Biomedical Engineering, Yale University, New Haven, CT 06520, USA Department of Mechanical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, 790-784, Korea Department of Plastic Surgery, College of Medicine, The Catholic University of Korea, Seoul, 137-701, Korea Department of Chemical Engineering, Pohang University of Science and Technology (POSTECH), San 31, Hyoja dong, Nam-gu, Pohang, Gyungbuk, 790-784, Korea Authors contributed equally to this study.
E-mail: [email protected], [email protected] and [email protected] Keywords: tissue engineering, vascular graft, endothelization, thrombosis Supplementary material for this article is available online
Abstract The use of tissue mimics in vivo, including patterned vascular networks, is expected to facilitate the regeneration of functional tissues and organs with large volumes. Maintaining patency of channels in contact with blood is an important issue in the development of a functional vascular network. Endothelium is the only known completely non-thrombogenic material; however, results from treatments to induce endothelialization are inconclusive. The present study was designed to evaluate the clinical applicability of in situ recruitment of endothelial cells/endothelial progenitor cells (EC/EPC) and preendothelization using a recombinant mussel adhesive protein fused with arginine–glycine–aspartic acid peptide (MAP-RGD) coating in a model of vascular graft implantation. Microporous polycaprolactone (PCL) scaffolds were fabricated with salt leaching methods and their surfaces were modified with collagen and MAP-RGD. We then evaluated their anti-thrombogenicity with an in vitro hemocompatibility assessment and a 4-week implantation in the rabbit carotid artery. We observed that MAP-RGD coating reduced the possibility of early in vivo graft failure and enhanced re-endothelization by in situ recruitment of EC/EPC (patency rate: 2/3), while endothelization prior to implantation aggravated the formation of thrombosis and/or IH (patency rate: 0/3). The results demonstrated that in situ recruitment of EC/EPC by MAP-RGD could be a promising strategy for vascular applications. In addition, it rules out several issues associated with pre-endothelization, such as cell source, purity, functional modulation and contamination. Further evaluation of long term performance and angiogenesis from the luminal surface may lead to the clinical use of MAP-RGD for tubular vascular grafts and regeneration of large-volume tissues with functional vascular networks.
Introduction Blood vasculature is viewed as a dynamic network that supplies tissues with nutrients and gases and regulates various processes through direct communication of endothelial cells (EC) with adjacent cells [1–3]. Recent advances in fabrication technology have enabled the formation of tissue mimics, including patterned vascular networks, by lining ECs in preconstructed © 2015 IOP Publishing Ltd
fluidic channels [4–8]. Studies of tissue mimics in vitro provide a better understanding of angiogenesis [5, 6, 9–12], cancer cell invasion/extravasation [9–11], and interactions between ECs and parenchymal cells [5, 9, 12]. Moreover, the use of the tissue mimics in vivo is expected to facilitate the regeneration of functional tissues and organs with large volumes, because surgical connection of the artificial vascular network to the host vasculature is the only tangible
Biofabrication 7 (2015) 015007
T-Y Kang et al
solution that will meet the metabolic needs of engineered tissue/organ constructs [13–15]. Therefore, maintaining patency of channels in contact with blood is important in developing a functional vascular network. Endothelium is the only known completely non-thrombogenic material, and previous studies have showed that the luminal endothelium prevents the activation of inflammatory and coagulatory events [16–19]. However, research in the field of small-diameter graft development found that successful seeding of EC does not necessarily result in a positive outcome clinically. Two main strategies are being explored to create endothelialized vascular grafts. The first is to seed cells in the graft and grow a confluent endothelial layer on the luminal surface before implantation. This technique has been studied with ECs from autologous veins [20] and adipose tissue [21, 22], or with endothelial progenitor cells (EPC) from bone marrow [23] and peripheral blood [24]. The second strategy is to induce endothelization in vivo by recruiting circulating EPCs through surface modifications with anti-CD34 antibodies [25] and RGD [26, 27]. Although some treatments enhanced endothelization and prevented early graft failure by thrombosis, they stimulated the formation of intimal hyperplasia (IH), especially at anastomotic sites, and induced EC and smooth muscle cell (SMC) dysfunction [22, 28, 29]. Our previous experience in vitro proved that the recombinant mussel adhesive protein fused with arginine–glycine–aspartic acid peptide (MAP-RGD) coating improved endothelization and inhibited platelet activation [19]. As the MAP-RGD coating can be easily applied to complex channels, the results suggested the potential of MAP-RGD for the development of a functional vascular network with channel patency. Therefore, we evaluated the clinical applicability of in situ recruitment of EC/EPC and pre-endothelization using a MAP-RGD coating in a model of vascular graft implantation.
4 mm by 8 mm pieces. To fabricate a tubular scaffold as a vascular graft, a 15-gauge needle was prepared as a mold; the outer diameter of this needle, 1.829 mm, corresponds to the inner diameter of the scaffold. The mold was dipped in the PCL mixture, placed in ambient air to extract chloroform, and then placed in deionized water to dissolve the sodium chloride particles. Finally, a tubular scaffold was prepared by taking off the needle and trimming the ends. The scaffold’s surfaces were modified with collagen and MAP-RGD as described previously [19]. Acid-soluble type I collagen (Koken atelocollagen implant, Koken, Japan) was dissolved in 0.1 M hydrochloric acid to a concentration of 1% (w/v). Scaffolds were immersed in collagen solution for 60 min at 4 °C for immobilization. The collagen-coated scaffolds were then crosslinked by using N-hydroxysuccinimide(NHS)/N-(3dimethylaminopropyl)-N’-ethylcarbodiimide (EDC) solution in ethanol (EDC/NHS at 1:1 each 10 mg ml−1 in 95% ethanol) for 3 h at 4 °C [34]. The remnant collagen was washed with distilled water three time. After crosslinking, the samples were freeze-dried and the process was repeated twice. For MAP immobilization, MAP-RGD solution was prepared by dissolving DOPA-modified fp-151-RGD in distilled water (DW) at 2 mg ml−1. Collagen-coated scaffolds were immersed in MAP-RGD solution for 24 h at room temperature then washed with phosphate buffered saline (PBS). A detailed description of MAP-RGD production and characteristics were previously reported [30–32]. For MAP immobilization, MAPRGD solution was prepared by dissolving 3,4-dihydroxyphenylalanine-modified fp-151-RGD in distilled water. We planned to adopt a collagen layer as a basement membrane on the luminal surface of an artificial vascular network in a porous scaffold. The collagen layer should be a barrier to maintain endothelial cells on the surface. Thus, the vascular grafts used in this study were also coated with collagen prior to MAP-RGD to provide the same environment as the artificial vascular network.
Materials and methods Scaffold fabrication and surface modification Thin films and tubular scaffolds were fabricated to evaluate the blood compatibility of surfaces in vitro and patency in vivo, respectively. A polycaprolactone (PCL; Polyscience, Inc., MW 43 000–50 000) solution was prepared in chloroform at a concentration of 10% (w/v). Milled sodium chloride particles (SigmaAldrich, USA) were sieved through a 45 μm mesh. These were then mixed with the PCL solution at a composition of 400 wt.% salt of PCL. To fabricate a porous and thin film, the PCL mixture was poured on a glass plate and dried in ambient air for 3 h. After dissolving the sodium chloride particles in water for 1 h, the film was peeled from the glass and cut into 2
Evaluation of mechanical properties Uniaxial tensile strength test was carried out to assess the effect of the amount of sodium chloride on graft mechanical properties using a test equipment (Instron, Norwood, MA, USA). The thin film specimens (25 mm × 12.5 mm × 300 μm) were prepared as described above and the sodium chloride particles were mixed with the PCL solution at a composition of 0, 100, 200, 400, and 600 wt.% salt of PCL. In the tensile test, the specimens were pulled at a crosshead velocity of 5 mm min−1. The mechanical tensile modulus was calculated from the stress-strain curves. In the stress-strain curve, the slope of the first linear portion was defined as the modulus of each specimen.
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 1. Perfusion system. Here, we show the configuration of the perfusion system (A), the perfusion chamber for the hemocompatibility test (B), and culture using tubular scaffolds (C).
In vitro hemocompatibility test with whole blood We employed a perfusion system (figure 1(A)) in an incubator with a 5% CO2 atmosphere at 37 °C to elucidate the interaction of blood and prepared samples. The system configuration was described in our previous study [19]. Briefly, the system was composed of a peristaltic pump, reservoir, gas exchanger, and a perfusion chamber designed to load thin films (figure 1(B)). Fresh blood was collected with the use of Vacutainer tubes containing sodium citrate (BD, USA) from a healthy adult volunteer free of aspirin or other drugs that could bias the results. Prior to blood circulation, all components of the perfusion system and samples were sterilized for 2 h with 70% ethanol and ultraviolet (UV) light. Any remaining ethanol was removed by washing with PBS. The flow rate was set to match physiological shear stress levels [33] (1 Pa) in a carotid artery of a rabbit. Whole blood was circulated for 3 h in the perfusion system, after which the scaffolds, in the form of thin films, were taken out and fixed in 10% formalin for 30 min. During the whole blood circulation for 3 h, blood wasn’t coagulated in the system. However, in our experience, it started to be coagulated after 8 h. Immunohistochemical staining was used to visualize the adhered platelets, leukocytes and EPCs using antibodies to CD41 (abcam, USA), CD45 (abcam, USA), and CD34 (biorbyt Ltd, UK), respectively. All images were obtained using a confocal microscope (Olympus confocal FluoView 1000). And the coverage ratio of platelets and the numbers of leukocytes/EPCs were quantified by imageJ program with3 images acquired from different sites. In an image, cells stained 3
with DAPI/ CD45 were counted as leukocyte but cells stained with DAPI/CD 34 were counted as EPC. Endothelial cell isolation and characterization ECs were isolated by following a protocol described by Russell et al [34]. A male rabbit (3.0 kg) was sacrificed by isoflurane inhalation. The thoracic aorta was excised immediately from the heart after sacrifice and rinsed in PBS. The aorta was transferred to a dish filled with a collagenase solution and incubated for 5–10 min at 37 °C. The endothelial layer was then gently removed by rotating a dry, sterile swab onto the surface of the aorta and dabbing the swab in the collagenase solution to dislodge cells from the tip. Cells were collected by centrifugation at 1000 rpm for 5 min and the supernatant was aspirated. After adding 3 mL of endothelial growth medium (EGM-2) with 10% fetal bovine serum (Thermo Scientific HyClone, Waltham, MA) and 1% antibiotic/antimycotic solution (Gibco, Carlsbad, CA), the cell suspension was centrifuged again. The cells were cultured in the same medium in an incubator with a 5% CO2 atmosphere at 37 °C, and used at passage 4 for all experiments. Immunohistochemical staining was used to identify the isolated cells with antibodies to von Willebrand factor (vWf, biorbyt Ltd, UK) and α-smooth muscle actin (α-SMA, abcam, USA). All images were obtained using a confocal microscope. Cell seeding and perfusion culture Segments of a silicone tube were prepared as a perfusion chamber for tubular scaffolds (figure 1(C)). After expansion, an EC suspension was injected into
Biofabrication 7 (2015) 015007
T-Y Kang et al
the perfusion chamber with fluorescent labeling (Neostem, South Korea), followed by shaking and incubating for 3 h. The chamber was then connected to the perfusion system in an incubator with a 5% CO2 atmosphere at 37 °C. Cell-seeded scaffolds were cultured under perfusion conditions for 6 days, while the shear stress was increased from 0.1 Pa to 1 Pa to maximize the proliferation rate, as shown in our previous study [19]. In vivo implantation and sacrifice The experimental protocol was approved by the Catholic University of Korea Animal Care and Use Committee. Fifteen New Zealand White rabbits (2.5–3.0 kg) were divided into 5 groups: bare PCL, bare PCL/Collagen, bare PCL/Collagen/MAP-RGD, cell-seeded PCL/Collagen, and cell-seeded PCl/Collagen/MAP-RGD. Under general anesthesia by intravenous injection of combination of intraperitoneal ketamine (75 mg kg−1) and xylazine (10 mg kg −1), the neck skin was shaved and the right carotid artery was exposed. The artery was clamped at the proximal and distal ends and then transected. The tubular scaffold was trimmed to 8mm in length and sutured end to end with 8 to 10 interrupted stiches using a 9-0 ethilon suture. No anticoagulation or anti-platelet drugs were given before or after the implantation procedure. After 4 weeks, the animals were reanesthetized and the patency of scaffolds was observed by restoration of blood flow in the distal vessel after squeezing and releasing the vessel with forceps. More specifically, one pair of forceps was used to hold the distal blood vessel near the anastomotic site, while the other pair of forceps was used to squeeze the vessel toward the distal end, so the vessel became flat. We then released the forceps near the anastomotic sites to observe that blood flow could be restored, inflating the vessel. Scaffolds were subsequently resected and washed with heparinized saline to remove the remaining blood.
Histological analysis Grafts were explanted after 4 weeks, fixed in 10% formalin for 30 min, and divided into proximal, middle, and distal regions. The pieces were separately embedded in O.C.T. compound (Tissue-Tek, Sakura Fineteck Inc., Torrance, CA, USA) and 4 μm cryosections were taken for histological evaluation. The proximal and distal regions were sectioned longitudinally, but the middle regions were sectioned transversely. All sections were removed from O.C.T. using 10% formalin and stained with hematoxylin and eosin (H&E). Immunohistochemical staining was used to analyze the sections with primary antibodies to vWf (biorbyt, UK) and α-SMA (abcam, USA). All images were obtained using a confocal microscope. 4
1.1. Statistical analysis Statistical analyses were performed using one-way analysis of variance (ANOVA) with a post-hoc Tukey test using MINITAB (version 14.2). Differences between groups were considered statistically significant at P < 0.05.
Results Microporous thin films and tubular scaffolds Thin films (figure 2(A)) and tubular scaffolds (figures 2(B) and (C)) were successfully fabricated using salt-leaching methods. As shown in figure 2(D), the scaffolds exhibited irregular pores on the surface. After collagen and MAP-RGD coating, the pores were covered by a thin layer of collagen (figure 2(E)) and the flat surface became grainy in sequence (figure 2(F)). Uniaxial tensile modulus according to the amount of sodium chloride Figure 2(G) shows the tensile modulus changes according to the amount of sodium chloride. The tensile strength of PCL specimen was about 235 MPa, which is much higher than those of natural vessels. However, microporous specimens fabricated by salt leaching method showed significantly decreased values. When the salt/polymer ratio was 4 (400%) to 6 (600%), the tensile modulus values were comparable to the values of the natural vessels [35–37]. By considering the ease of handling for the microsurgical anastomosis, 400 wt.% salt of PCL was used for the in vivo implantation. Hemocompatibility of bare PCL, PCL/Collagen, and PCL/Collagen/MAP-RGD The contact of blood with biomaterials triggers the coagulation and complement cascades, resulting in thrombosis and inflammation [38]. The ideal biomaterial for vascular applications exhibits low thrombogenicity and rapid re-endothelization when in contact with circulating blood. Therefore, we characterized the hemocompatibility of the samples based on the adhesion of platelets (figures 3(A)–(C)), leukocytes (figures 3(D)–(F)) and EPCs (figures 3(G)-(I)). As shown in figure 3(J), there was no statistically significant difference in the number of adhered platelets on the PCL, PCL/Collagen, and PCL/Collagen/MAPRGD grafts. However, PCL/Collagen/MAP-RGD grafts had fewer leukocytes (figure 3(K)), but markedly more EPCs (figure 3(L)) on the surface than PCL and PCL/Collagen grafts. Characteristics of isolated endothelial cells Figure 4(A) shows the morphology of the cells isolated form a heart artery and passaged up to four times. Some of the cells grew in a ‘cobblestone morphology’, which is characteristic of EC. However, some grew in a
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 2. Structural characterization of thin films and grafts. Optical image of fabricated PCL thin film (A) and tubular grafts (B); scanning electron microscopy image for the cross-section the graft (C); luminal surface of PCL (D), PCL/Collagen (E), and PCL/ Collagen/MAP-RGD grafts (F); (G) tensile modulus of thin films according to the amount of sodium chloride.
‘spindle-shaped’ pattern, which is the characteristic of SMC. To confirm that the isolated cells were contaminated by SMCs, double-staining was carried out with vWf (biorbyt Ltd, UK) and α-SMA (abcam, USA), which are specific markers for EC and SMC, respectively. Figure 4(B) indicates that SMCs were still present with ECs even at passage 4. In vivo patency of implanted grafts To compare the anti-thrombogenic properties of vascular grafts in vivo, we implanted bare PCL, bare PCL/Collagen, bare PCL/Collagen/MAP-RGD grafts, cell-seeded PCL/Collagen, and cell-seeded PCL/Collagen/MAP-RGD grafts into the common carotid artery of rabbits for 4 weeks. Patency was determined by restoration of blood flow in the distal vessel after squeezing and releasing the vessel with forceps (Movie S1-S2) and histological analysis (figures 5(C)–(F)). Among bare grafts, 3 of 3 PCL and 2 of 3 PCL/ Collagen/MAP-RGD grafts were patent. As one of the PCL/Collagen/MAP-RGD grafts was broken, the 5
patency could not be observed. The failed PCL/ Collagen/MAP-RGD graft was found to be broken into pieces and surrounded by fibrosis. All bare PCL/ Collagen, cell-seeded PCL/Collagen/MAP-RGD, cellseeded PCL/Collagen/MAP-RGD grafts showed no restoration of blood flow and were therefore considered to be completely occluded. H&E images of bare PCL and PCL/Collagen/MAP-RGD grafts show that the middle region and proximal/distal anastomotic sites remained patent with little neointimal formation. On the other hand, the occluded bare PCL/Collagen, cell-seeded PCL/Collagen, and cell-seeded PCL/Collagen/MAP-RGD grafts were clogged due to severe thrombosis and IH. In vivo endothelization and intimal hyperplasia Immunofluorescent staining revealed signs of IH and/ or endothelization on the luminal walls of the grafts (figures 5(A)–(F)). The patent grafts (bare PCL and bare PCL/Collagen/MAP-RGD) showed different results. The bare PCL grafts had evenly distributed IH
Biofabrication 7 (2015) 015007
T-Y Kang et al
Coverage ratio [%]
J
Cell Number
K
Cell Number
L
18 16 14 12 10 8 6 4 2 0
45 40 35 30 25 20 15 10 5 0
Platelet Adhesion
PCL
PCL/Collagen
PCL/Collagen/MAP-RGD
Leukocyte adhesion
PCL
PCL/Collagen PCL/Collagen/MAP-RGD
EPC adhesion
60 50 40 30 20 10 0
PCL
PCL/Collagen PCL/Collagen/MAP-RGD
Figure 3. Blood-material interaction in vitro. Platelets adhered to PCL (A), PCL/Collagen (B), and PCL/Collagen/MAP-RGD (C) are in green, labeled with CD41. Leukocytes adhered on PCL (D), PCL/Collagen (E), PCL/Collagen/MAP-RGD (F) are in red, labeled with CD45. EPCs adhered on PCL (G), PCL/Collagen (H), PCL/Collagen/MAP-RGD (I) are in green, labeled with CD34. We quantified adhered platelets (J), leukocytes (K), and EPCs (L) from the images.
Figure 4. Characterization of isolated ECs showed the microscopic appearance of isolated cells at passage 4 (A), ECs staining positive for the von Willebrand factor, and smooth muscle cells staining positive for α-smooth muscle actin (B).
that was stained with α-SMA, but we could not observe regenerated endothelium in the lumen (figure 6(A)). On the other hand, the bare PCL/Collagen/MAPRGD had non-uniform IH, which was rarely present in one region, but thicker than that of bare PCL grafts in another region. Interestingly, we observed endothelial regeneration in the lumen even though endothelial coverage was not complete (figure 6(C)). The occluded grafts (bare PCL/Collagen, cell-seeded PCL/ Collagen, and cell-seeded PCL/Collagen/MAP-RGD) were filled with endothelial cells and smooth muscle cells. Here, we present an image of a bare PCL/ Collagen graft (figure 6(B)) as a representative occluded graft. Before implantation, the cell-seeded grafts were completely covered with primary 6
endothelial cells, as shown in figure 6(D). However, the fluorescently (Neo-stem, South Korea) labeled cells were present in the occluded region of the cellseeded PCL/Collagen/MAP-RGD grafts (figure 6(F)), but not in cell-seeded PCL/Collagen grafts (figure 6(E)). The results indicated that the seeded cells were retained by MAP-RGD, but they actually aggravated the formation of thrombosis and/or IH.
Discussion The unique capacity of EC as an anti-thrombotic surface renders re-endothelization to be an effective approach for vascular applications. However, previous research has shown that vascular biology is more
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 5. An image taken immediately after implantation of a graft (A); illustration of graft regions for histological analysis (B); representative H&E staining images of bare PCL (C), bare PCL/Collagen (D), bare PCL/Collagen/MAP-RGD (E), cell-seeded PCL/ Collagen (F), and cell-seeded PCL/Collagen/MAP-RGD (G). The blue arrows indicate the implanted graft structures. The tissues formed inside of the grafts represent the intimal hyperplasia.
7
Biofabrication 7 (2015) 015007
T-Y Kang et al
Figure 6. Characterization of the vascular neotissues by immunofluorescent staining. Formations of the smooth muscle layer and endothelial layer stained with α-SMA and vWf, respectively, on bare PCL (A), bare PCL/Collagen (B), and bare PCL/Collagen/MAPRGD (C) grafts, the thickness of SMC layer represents the extent of SMC invasion. The bare PCL/Collagen/MAP-RGD had nonuniform IH and a newly formed endothelial layer; cells covering the inner surface of a graft before implantation (D); absence of green labeled cells indicating the loss of in vitro seeded cells in a PCL/Collagen graft (E); presence of green labeled cells indicating the retention of in vitro seeded cells in a PCL/Collagen/MAP-RGD graft (F); white dotted lines indicate the interface of microporous PCL part and thrombosis/IH.
complex than originally believed, and it is still a challenge to enhance the formation of endothelium while preventing the activation of inflammatory/ thrombotic events and abnormal IH [22, 23, 28]. In this regard, surface modification with MAP-RGD is a valuable strategy to improve graft patency. In vitro hemocompatibility assessment of bare PCL, PCL/Collagen, and PCL/Collagen/MAP-RGD revealed several interesting points that influenced the in vivo vascular patency and organization. The interaction of bare PCL/Collagen/MAP-RGD with blood was compared with the results from bare PCL and bare PCL/Collagen. PCL is a well-known biocompatible polymer [39] and it is an adequate material for vascular applications because of its excellent mechanical properties and slow degradation rate in vivo. Although vascular grafts made of PCL showed a promising patency rate [39, 40], modifications of PCL grafts have been attempted to enhance the adhesion of vascular cells for long term performance [41–43]. Collagen, a natural extracellular matrix polymer, is favorable for cell adhesion, but it is a highly thrombogenic material [44]. Bare PCL/Collagen/MAP-RGD had better hematocompatibility than the other two groups. Bare PCL/Collagen/MAP-RGD also had fewer leukocytes but more EPCs on the surface than bare PCL or bare PCL/Collagen. Although there was no statistical difference in platelet adhesion, we previously found that 8
platelets were not activated on PCL/Collagen/MAPRGD [19]. Consistently, in vivo studies showed that a MAPRGD coating reduced the possibility of early in vivo graft failure and enhanced re-endothelization by in situ recruitment of EC/EPCs. The patency rate of bare PCL/Collagen/MAP-RGD grafts (2/3) was lower than that of bare PCL grafts (3/3) because the exceptionally broken PCL/Collagen/MAP-RGD graft was considered as a failure. However, histological staining showed that bare PCL/Collagen/MAP-RGD had more potential as a graft in vascular regeneration than bare PCL. During its implantation, the animal slightly awoke from the anesthesia and moved over. We assumed that the graft had been damaged at that moment and broken in the end. Although we followed a standard protocol of anesthesia, there was still an individual difference for each animal. Therefore, more delicate efforts will be required to adjust the anesthetic drugs. After the implantation of vascular grafts in vivo, the primary change seen in the lumen is IH, which reflects the accumulation of inflammatory, coagulatory, and hemodynamic factors [25, 45–47]. IH consists predominantly of smooth muscle cells, and excessive IH usually leads to graft failure [28, 48]. Bare PCL grafts showed spatially homogeneous IH at the middle region and proximal/distal anastomotic sites. However, bare PCL/Collagen/MAP-RGD grafts showed less IH than PCL at proximal and distal
Biofabrication 7 (2015) 015007
T-Y Kang et al
anastomotic sites, where IH usually develops significantly more than in the graft body [49]. As shown in figure2(D), MAP-RGD doesn’t cover the whole surface. In our previous research [19], on the whole, the flat surface had become grainy by MAP-RGD. Compared with the flat sample used in the previous research, more efforts is required to coat MAP-RGD on the surface of the tubular graft. The spatially nonhomogeneous IH formation at the middle region of the PCL/Collagen/MAP-RGD graft might be caused by the nonhomogeneous MAP-RGD coating. We expect that the IH formation at the middle region could be reduced by increasing the number of MAPRGD coatings prior to implantation. Moreover, MAPRGD has been proven to accelerate re-endothelization, which is a crucial factor in maintaining the long term patency of a graft. Progress of EC coverage was found even at the middle region, remote from anastomotic sites of bare PCL/Collagen/MAP-RGD graft, whereas there was no EC coverage at the same region of bare PCL grafts. And the extent of SMC invasion was inhibited by the newly formed endothelium on PCL/Collagen/MAP-RGD graft. Re-endothelization of a graft could occur via out-growth of local endothelial cells from the host vessels adjacent to anastomotic sites or via recruitment of circulating endothelial progenitor cells as the re-endothelization of injured arteries [50]. MAP-RGD coating could have positive influences on both of mechanisms of re-endothelization. Previous studies have reported the capacity of RGD to promote the attachment of ECs [27, 43], capture circulating EPCs [51], and inhibit platelet adhesion/activation [19, 43]. In addition, our present study showed that MAP-RGD reduced the adhesion of leukocytes in blood. Based on these results, the present in vitro and in vivo study demonstrated that the fusion of RGD with MAP did not hamper intrinsic hemocompatibility while improving the immobility of RGD. If MAP-RGD coating is applied to the construction of a functional vascular network, EC/EPC recruitment by MAP-RGD could accelerate angiogenesis not only from the luminal surface of a vascular network, but also from the outside of the scaffold. Here, we observed that endothelization prior to implantation aggravated the formation of thrombosis and/or IH. Using a fluorescent dye, it has been shown that after 4 weeks of follow-up, seeded cells remained in PCL/Collagen/MAP-RGD grafts due to the strong binding affinity of MAP-RGD, but not in PCL/Collagen. However, the retained cells were found in the occluded region and it is suggested that the cells contributed to thrombosis and/or IH. The failure of the cell-seeded graft could be caused by complicated factors occurring in any step of the procedures for harvest, expansion, and perfusion culture. Although we carefully swabbed the luminal surface of the vessel to harvest only ECs, a number of SMCs were present and we seeded cells without purification. Many researchers have tried to develop effective techniques to isolate 9
and purify endothelial cells. However, our study was designed to evaluate the feasibility of MAP-RGD for in situ recruitment of endothelial cells and compare it with risky environments by pre-seeded primary cells. Although it will take more time to purify and expand cells, delicately purified cells could increase the patency of vascular grafts. Second, the exposure of ECs to media containing fetal bovine serum possesses the potential for both genotypic and phenotypic modulation of EC during in vitro expansion and perfusion culture. In addition, the human manipulation of the perfusion culture system could introduce the risk of contamination. We actually went through contamination during the initial experiments. Perfusion system consists of several components and it has high risk of contamination in each step such as assembling the whole system, transferring it from a bench to incubation. At the initial stage, we made efforts to get rid of unnecessary steps and ensure the sterilized status. The present study was designed to evaluate the clinical applicability of MAP-RGD coating in the construction of a functional vascular network for large tissue regeneration. Therefore, the graft patency was observed for a relatively short term with the assumption that blood delivery by the vascular network to the interior could be replaced by angiogenesis in the scaffold during the period. However, we cannot guarantee the patency of the vascular network in the same period of this work, because a vascular network could have smaller and more complex geometry than a simple vascular graft. Moreover, long term graft performance such as patency, structural integrity, calcification, polymer degradation, and vascular remodeling needs to be evaluated to create clinically acceptable vascular prostheses. Especially, the mismatch of mechanical properties between an engineered graft and native artery could cause excessive intimal hyperplasia of the connected artery [52]. Although the data were not presented here, in our initial experiment, the native vessels, which were connected to relatively rigid tubular grafts, were found to be not only occluded but also hardened. Further investigation will be required to elucidate the effect of mechanical compliance on the graft patency and integrity to the native vessel.
Conclusion This study evaluated the clinical applicability of a MAP-RGD coating in a model of vascular graft implantation. Here, we observed that MAP-RGD coating reduced the possibility of early in vivo graft failure and enhanced re-endothelization by in situ recruitment of EC/EPC (patency rate: 2/3), while endothelization prior to implantation aggravated the formation of thrombosis and/or IH (patency rate: 0/ 3). The results demonstrated that in situ recruitment of EC/EPC by MAP-RGD could be a promising strategy for vascular application that rules out several
Biofabrication 7 (2015) 015007
T-Y Kang et al
issues of pre-endothelization such as cell source, purity, functional modulation, and contamination. Further evaluation of the long term performance and angiogenesis from the luminal surface may potentially lead to the clinical use of MAP-RGD not just for tubular vascular grafts, but for regeneration of large volume tissues with functional vasculature.
[16]
[17]
[18]
Acknowledgments This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (No. 2012-0001235 and No. 2011-0030075) (to D.-W. C), Marine Biotechnology Program funded by the Ministry of Oceans and Fisheries, Korea (to H.J.C) and the Rising Star Program funded by POSTECH (to H.J.C).
[19]
[20]
[21]
References [1] Cines D B et al 1998 Endothelial cells in physiology and in the pathophysiology of vascular disorders Blood 91 3527–61 [2] Cleaver O and Melton D A 2003 Endothelial signaling during development Nat. Med. 9 661–8 [3] Lazarus A, Del-Moral P M, Ilovich O, Mishani E, Warburton D and Keshet E 2011 A perfusion-independent role of blood vessels in determining branching stereotypy of lung airways Development 138 2359–68 [4] Choi N W, Cabodi M, Held B, Gleghorn J P, Bonassar L J and Stroock A D 2007 Microfluidic scaffolds for tissue engineering Nat. Mater. 6 908–15 [5] Miller J S et al 2012 Rapid casting of patterned vascular networks for perfusable engineered three-dimensional tissues Nat. Mater. 11 768–74 [6] Zheng Y et al 2012 In vitro microvessels for the study of angiogenesis and thrombosis Proc. Natl. Acad. Sci. USA 109 9342–7 [7] Jung J W, Kang H W, Kang T Y, Park J H, Park J and Cho D W 2012 Projection image-generation algorithm for fabrication of a complex structure using projection-based microstereolithography Int. J. Precis. Eng. Man. 13 445–9 [8] Seol Y J, Kang T Y and Cho D W 2012 Solid freeform fabrication technology applied to tissue engineering with various biomaterials Soft. Matter 8 1730–5 [9] Chung S, Sudo R, Vickerman V, Zervantonakis I K and Kamm R D 2010 Microfluidic platforms for studies of angiogenesis, cell migration, and cell-cell interactions Ann. Biomed. Eng. 38 1164–77 [10] Zervantonakis I K, Hughes-Alford S K, Charest J L, Condeelis J S, Gertler F B and Kamm R D 2012 Threedimensional microfluidic model for tumor cell intravasation and endothelial barrier function Proc. Natl. Acad. Sci. USA 109 13515–20 [11] Jeon J S, Zervantonakis I K, Chung S, Kamm R D and Charest J L 2013 In vitro model of tumor cell extravasation PLOS One 8 e56910 [12] Huh D, Matthews B D, Mammoto A, Montoya-Zavala M, Hsin H Y and Ingber D E 2010 Reconstituting organ-level lung functions on a chip Science 328 1662–8 [13] Baptista P M, Siddiqui M M, Lozier G, Rodriguez S R, Atala A and Soker S 2011 The use of whole organ decellularization for the generation of a vascularized liver organoid Hepatology 53 604–17 [14] Arenas-Herrera J E, Ko I K, Atala A and Yoo J J 2013 Decellularization for whole organ bioengineering Biomed. Mater. 8 014106 [15] Mirmalek-Sani S H, Deegan D B, Sullivan D C, Ko I, George S K, Jackson J D, Aboushwareb T A, Atala A and Yoo J
10
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29] [30]
[31]
[32]
[33]
[34] [35]
2012 Optimizing decellularization of porcine kidneys for whole organ bioengineering J. Tissue Eng. Regen. M 6 172 Blit P H, McClung W G, Brash J L, Woodhouse K A and Santerre J P 2011 Platelet inhibition and endothelial cell adhesion on elastin-like polypeptide surface modified materials Biomaterials 32 5790–800 Knetsch M L W, Aldenhoff Y B J, Schraven M and Koole L H 2004 Human endothelial cell attachment and proliferation on a novel vascular graft prototype J. Biomed. Mater. Res. A 71A 615–24 Wu H C, Wang T W, Kang P L, Tsuang Y H, Sun J S and Lin F H 2007 Coculture of endothelial and smooth muscle cells on a collagen membrane in the development of a smalldiameter vascular graft Biomaterials 28 1385–92 Kang T Y, Hong J M, Kim B J, Cha H J and Cho D W 2013 Enhanced endothelialization for developing artificial vascular networks with a natural vessel mimicking the luminal surface in scaffolds Acta Biomater. 9 4716–25 Lamm P, Juchem G, Milz S, Schuffenhauer M and Reichart B 2001 Autologous endothelialized vein allograft—a solution in the search for small-caliber grafts in coronary artery bypass graft operations Circulation 104 I108–14 Koyama M, Satoh K, Yoshida H, Suzuki S, Koie H and Takamatsu S 1996 Surface coverage of vascular grafts with cultured human endothelial cells from subcutaneous fat tissue obtained with a biopsy needle Thromb. Haemostasis. 76 610–4 Arts C H P, Joosten P P A H, Blankensteijn J D, Staal F J T, Ng P Y Y, Heijnen-Snyder G J, Sixma J J, Verhagen H J M, de Groot P G and Eikelboom B C 2002 Contaminants from the transplant contribute to intimal hyperplasia associated with microvascular endothelial cell seeding Eur. J. Vasc. Endovasc. 23 29–38 Jobst B J, Riegger G A J and Griese D P 2009 Endothelial cell seeding fails to prevent intimal hyperplasia following arterial injury in the rat carotid model Cardiovasc Drug Ther. 23 343–53 Brunt K R, Hall S, Ward C A and Melo L G 2007 Endothelial progenitor cell and mesenchymal stem cell isolation, characterization, viral transduction Methods Mol. Med. 139 197–210 Rotmans J I, Heyligers J M M, Verhagen H J M, Velema E, Nagtegaal M M, de Kleijn D P V, de Groot F G, Stroes E S G and Pasterkamp G 2005 In vivo cell seeding with anti-cd34 antibodies successfully accelerates endothelialization but stimulates intimal hyperplasia in porcine arteriovenous expanded polytetrafluoroethylene grafts Circulation 112 12–8 Li J H, Ding M M, Fu Q, Tan H, Xie X Y and Zhong Y P 2008 A novel strategy to graft rgd peptide on biomaterials surfaces for endothelization of small-diamater vascular grafts and tissue engineering blood vessel J. Mater. Sci-Mater. M 19 2595–603 Tugulu S, Silacci P, Stergiopulos N and Klok H A 2007 Rgd— functionalized polymer brushes as substrates for the integrin specific adhesion of human umbilical vein endothelial cells Biomaterials 28 2536–46 Subbotin V M 2007 Analysis of arterial intimal hyperplasia: review and hypothesis Theor. Biol. Med. Model 4 Loscalzo J 2000 Vascular matrix and vein graft failure—is the message in the medium? Circulation 101 221–3 Choi B H, Choi Y S, Kane D G, Kim B J, Song Y H and Cha H J 2010 Cell behavior on extracellular matrix mimic materials based on mussel adhesive protein fused with functional peptides Biomaterials 31 8980–8 Hwang D S, Gim Y, Yoo H J and Cha H J 2007 Practical recombinant hybrid mussel bloadhesive fp-151 Biomaterials 28 3560–8 Hwang D S, Sim S B and Cha H J 2007 Cell adhesion biomaterial based on mussel adhesive protein fused with rgd peptide Biomaterials 28 4039–46 Matsumoto T, Okumura E, Miura Y and Sato M 1999 Mechanical and dimensional adaptation of rabbit carotid artery cultured in vitro Med. Biol. Eng. Comput. 37 252–6 Gould R A and Butcher J T 2010 Isolation of valvular endothelial cells J. Visualized Exp. 46 e2158 Williams C, Liao J, Joyce E M, Wang B, Leach J B, Sacks M S and Wong J Y 2009 Altered structural and
Biofabrication 7 (2015) 015007
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
T-Y Kang et al
mechanical properties in decellularized rabbit carotid arteries Acta Biomater. 5 993–1005 Hashi C K, Derugin N, Janairo R R, Lee R, Schultz D, Lotz J and Li S 2010 Antithrombogenic modification of small-diameter microfibrous vascular grafts Arteriosclerosis, Thrombosis, And Vascular Biology 30 1621–7 Hasan A, Memic A, Annabi N, Hossain M, Paul A, Dokmeci M, Dehghani F and Khademhosseini A 2014 Electrospun scaffolds for tissue engineering of vascular grafts Acta Biomater. 10 11–25 Gorbet M B and Sefton M V 2004 Biomaterial-associated thrombosis: Roles of coagulation factors, complement, platelets and leukocytes Biomaterials 25 5681–703 de Valence S, Tille J C, Mugnai D, Mrowczynski W, Gurny R, Moller M and Walpoth B H 2012 Long term performance of polycaprolactone vascular grafts in a rat abdominal aorta replacement model Biomaterials 33 38–47 Mathews A, Colombus S, Krishnan V K and Krishnan L K 2012 Vascular tissue construction on poly(e-caprolactone) scaffolds by dynamic endothelial cell seeding: effect of pore size J. Tissue Eng. Regen. M 6 451–61 Zhang M, Wang K, Wang Z X, Xing B, Zhao Q and Kong D L 2012 Small-diameter tissue engineered vascular graft made of electrospun pcl/lecithin blend J. Mater. Sci-Mater. M 23 2639–48 Wissink M J B, Beernink R, Poot A A, Engbers G H M, Beugeling T, van Aken W G and Feijen J 2000 Improved endothelialization of vascular grafts by local release of growth factor from heparinized collagen matrices J. Control Release 64 103–14 Zheng W T et al 2012 Endothelialization and patency of rgdfunctionalized vascular grafts in a rabbit carotid artery model Biomaterials 33 2880–91
11
[44] Cloft H J et al 2000 Bovine type i collagen as an endovascular stent-graft material: biocompatibility study in rabbits Radiology 214 557–62 [45] Mattana J, Effiong C, Kapasi A and Singhal P C 1997 Leukocyte-polytetrafluoroethylene interaction enhances proliferation of vascular smooth muscle cells via tumor necrosis factor-alpha secretion Kidney Int. 52 1478–85 [46] Ojha M 1997 In vitro studies on anastomotic intimal hyperplasia Asaio J. 43 242 [47] Roy-Chaudhury P, Kelly B S, Miller M A, Reaves A, Armstrong J, Nanayakkara N and Heffelfinger S C 2001 Venous neointimal hyperplasia in polytetrafluoroethylene dialysis grafts Kidney Int. 59 2325–34 [48] Desai M, Mirzay-Razzaz J, von Delft D, Sarkar S, Hamilton G and Seifalian A M 2010 Inhibition of neointimal formation and hyperplasia in vein grafts by external stent/ sheath Vasc Med. 15 287–97 [49] Ojha M, Leask R L, Johnston K W, David T E and Butany J 2000 Histology and morphology of 59 internal thoracic artery grafts and their distal anastomoses Ann. Thorac. Surg. 70 1338–44 [50] Van der Heiden K, Gijsen F J, Narracott A, Hsiao S, Halliday I, Gunn J, Wentzel J J and Evans P C 2013 The effects of stenting on shear stress: relevance to endothelial injury and repair Cardiovasc Res. 99 269–75 [51] Kim K L et al 2009 Enhanced dermal wound neovascularization by targeted delivery of endothelial progenitor cells using an rgd-g-plla scaffold Biomaterials 30 3742–8 [52] Seifu D G, Purnama A, Mequanint K and Mantovani D 2013 Small-diameter vascular tissue engineering Nat. Rev. Cardiology 10 410–21
