Tissue engineering of blood vessel

Abstract

Vascular grafts are in large demand for coronary and peripheral bypass surgeries. Although synthetic grafts have been developed, replacement of vessels with purely synthetic polymeric conduits often leads to the failure of such graft, especially in the grafts less than 6 mm in diameter or in the areas of low blood flow, mainly due to the early formation of thrombosis. Moreover, the commonly used materials lack growth potential, and long-term results have revealed several material-related failures, such as stenosis, thromboembolization, calcium deposition and infection. Tissue engineering has become a promising approach for generating a bio-compatible vessel graft with growth potential. Since the first success of constructing blood vessels with collagen and cultured vascular cells by Weinberg and Bell, there has been considerable progress in the area of vessel engineering. To date, tissue- engineered blood vessels (TEBVs) could be successfully constructed in vitro, and be used to repair the vascular defects in animal models. This review describes the major progress in the field, including the seeding cell sources, the biodegradable scaffolds, the construction technologies, as well as the encouraging achievements in clinical applications. The remaining challenges are also discussed.

Introduction

Cardiovascular disease remains the leading cause of death in western countries and often requires vascular reconstruction. Autologous arteries or veins are the most commonly used substitutes for coronary and peripheral bypass procedures. However, autologous vessel is not available in over 10% of the patients as a result of trauma, vessel disease or previous surgery [1]. Early attempts to develop blood vessel substitutes have focused on the use of grafts engineered from synthetic material, such as ePTFE (expanded polytetrafluoroethylene) and Dacron (polyethylene terephthalate fibre). However, replacement of vessels with purely synthetic polymeric conduits often leads to the failure of such graft, especially in the small diameter (less than 6 mm) grafts or in the areas of low blood flow, mainly due to the early formation of thrombosis [2–4]. In addition, the materials that are commonly used lack growth potential and long-term results have revealed several material-related failures, such as stenosis, thromboembolization, calcium deposition and infection [5]. To solve these problems, in particular for children who require implantation of dynamic material with growth potential, optimal grafts with biocompatibility and growth potential are desirable.

By seeding functional cells on biodegradable scaffolds, tissue engineering has become a new approach for tissue regeneration [6]. Since Weinberg and Bell reported the construction of blood vessels with collagen and cultured bovine aortic endothelial cells, smooth muscle cells and adventitial fibroblasts in vitro[7], there has been considerable progress in the area of vascular engineering. To date, tissue-engineered blood vessels (TEBVs) could be successfully constructed in vitro, and be used to repair the vascular defects in animal models. However, only a few have achieved clinical success with this approach. This review aims to describe the major progress in the field, including the seeding cell sources, the biodegradable scaffolds, the construction technologies, as well as the encouraging achievements in clinical applications. The remaining problems will also be discussed in an effort to guide future endeavours.

Components of blood vessel

Blood vessels are made of three layers, called from the luminal side outward, the tunica intima, the tunica media and the tunica adventitia. The thickness of these three layers varies greatly depending upon the size and type of vessel (large, medium & small arteries and veins; capillaries, don't have three layers). The vascular wall (except for capillary), with its complicated architecture and unique mechanical properties, is mainly composed by three types of cells: the endothelial cells (ECs) that lined in the tunica intima, the smooth muscle cells (SMCs) that predominantly located in the tunica media and the adventitial fibroblasts in the tunica adventitia. Among them, ECs and SMCs play a pivotal role in keeping the integrity of the vessel and maintaining its mechanical properties. The endothelium layer provides a continuous selective permeable, thrombo-resistant barrier that facilitates laminar blood flow through the blood vessel. It also controls vessel tone, platelet activation, adhesion and aggregation, leukocyte adhesion and SMCs migration and proliferation. Meanwhile, SMCs have secretory capabilities. The collagen fibres, elastic fibres, elastic lamellae and proteoglycans secreted by the SMCs keep the elasticity and radial compliance of the vessel.

Principle of vessel engineering

The general approach of tissue engineering is to seed cells on biodegradable scaffolds first, followed by in vitro culture or in vivo implantation. Ideally, the scaffolds will be gradually resorbed, leaving only the new tissue generated by the cells. Thus, the successful tissue regeneration relies on the seeding cells, the scaffolds and the construction technologies [8, 9]. Functional TEBVs should be non-thrombogenic, non-immunogenic, compatible at high blood flow rates and have similar viscoelasticity to native vessels [10–12]. Moreover, the grafts should be living tissues that could eventually integrate into the body and become indistinguishable from the native vessels. It has been accepted that the functional TEBVs cannot be achieved without ECs, SMCs, biodegradable scaffolds and the unique vessel-engineering techniques (Fig. 1).

1.

Schematic diagram of engineering blood vessels by tissue-engineering approach for clinical application.

Seeding cell sources

The ideal cell source should be non-immunogenic, functional and easy to achieve and expand in culture. Mature vascular cells, embryonic and adult stem cells, as well as alternative cell types that could possibly replace the ECs and SMCs, have been testified in vessel engineering.

Autologous ECs and SMCs

Non-immunogenic autologous ECs and SMCs isolated from patients themselves are the first choice for vessel engineering. Cells isolated from autologous vessels have been well used for engineering new vessels by many groups [7, 13–15]. In 1986, Weinberg and Bell first constructed TEBVs with cultured bovine aortic ECs, SMCs and adventitial fibroblasts [7]. In our early study, we have performed similar research utilizing ECs and SMCs derived from canine carotid arteries or human umbilical veins (HUV) [16]. Although functional TEBVs could be constructed by seeding those cells on biodegradable scaffolds, the limited proliferation potential of harvested cells makes it impossible to obtain large amount of cells from a small vessel biopsy. It is known that the majority of the cells in adult blood vessel are terminally differentiated. Even the cells isolated from umbilical veins have limited proliferation potential [16]. In addition, cells would lose their function during in vitro expansion. Although Grenier et al. reported that ECs, SMCs and fibroblasts could be isolated simultaneously and expanded in culture from a single and small vein biopsy sample [17], the quality of the cells after expansion were not clear. Many attempts have been tried to improve the proliferation potential of ECs and SMCs. Genetic manipulation is one of the ways that have been tested. Mckee et al. introduced human telomerase reverse transcriptase subunit (hTERT) into human SMCs [18], while Shao et al. utilized the same approach to immortalize the primary human microvascular ECs [19]. Encouraging results approved that the resulting cells could proliferate far beyond their normal lifespan and retained their characteristics of normal control cells. However, the safety of the cells after genetic manipulation is still a great concern. Long-term follow-up of modified cells in vivo is necessary before application of those cells in clinic. Allogeneic ECs and SMCs is another source for vessel engineering. However, immuno-rejection problem could not be avoided in this case, especially for ECs that contact directly with blood cells. To date, there is no promising way to solve the cell proliferation problem. It is of great interest to find alternative cell sources for vessel engineering.

Embryonic stem cells

In the recent few years, stem cell has become a major cell source for tissue engineering [20–22]. Generally there are two types of stem cells based on their origin, the embryonic and adult stem cells. Embryonic stem (ES) cells are able to produce all types of cells, while adult stem cells are normally limited to certain lineages. The merit of utilizing stem cell as a seeding cell source is that those cells are able to self-renew and differentiate into mature cells in the proper conditions, which makes it possible to obtain large amount of functional cells for tissue regeneration.

Differentiation of ES cells into ECs and SMCs has been studied extensively in murine ES cells, including maturation steps, molecular events and growth factor involvement [23–26]. The foetal liver kinase-1 (Flk-1) positive cells from differentiated ES cells, containing EC and SMC progenitors, could participate the neovascular formation when injected into animal bodies [27]. In our early study, we have successfully induced mouse ES cells to differentiate into ECs, and those ECs were further immortalized by transfection with hTERT [28]. The immortalized cells were able to maintain the phenotype of normal ECs, including the expression of Flk-1, von Willebrand factor (vWF) and CD34. Cells could form tubular structures in the presence of vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF) and erythropoietin (EPO). Furthermore, we constructed a blood vessel by using SMCs obtained from rabbit arteries and the ECs derived from ES cells. This is the first work demonstrating that ES cells derived ECs could be a seeding cell source for vessel engineering. Recently, McCloskey et al. demonstrated that high purity of functional ECs could be achieved from differentiated mouse ES cells without genetic manipulation [29]. Moreover, Levenberg et al. showed that human ES cells could be differentiated into ECs that are able to form tube-like structures on matrigel, and form microvessels when they were transplanted into severe combined immune deficiency (SCID) mice [30]. All these achievements support that ES cells could be a good seeding cell source for vessel engineering. However, ES cells are still far away from their clinical application. Besides the ethical issue, immunogenic and tumourgenic problems are still the major obstacles that should be overcome before transplantation of those cells into the body.

Adult stem cells

Comparing with ES cells, adult stem cells can be obtained from patients themselves, which can avoid the immuno-rejection and ethical problems. In addition, adult stem cells are normally limited to certain lineages, which do not have tumorgenic capacity.

Endothelial progenitors cells (EPCs) are one type of the adult stem cells that have the capacity to proliferate, migrate and differentiate into mature ECs [31]. EPCs are mainly located in bone marrow and could be mobilized into peripheral blood by certain growth factors, such as granulocyte macrophage colony stimulating factor (GM-CSF) or VEGF [32–34]. EPCs could be also isolated from umbilical cord blood [35]. Studies have shown that EPCs are able to differentiate into mature ECs which express CD31, vWF and uptake low-density lipoprotein in the presence of VEGF [31]. In addition, EPCs could be expanded for over 20 passages without losing their differentiate potential [31]. No significant differences have been found between EPCs derived from bone marrow, peripheral blood or cord blood in terms of cell proliferation and differentiation [36]. EPCs have been well utilized in the endothelialization of synthetic vessel grafts as well as in vessel engineering [37–39]. Kaushal et al. isolated EPCs from peripheral blood of sheep, expanded and seeded them on decellularized porcine iliac vessels to construct an engineered vascular graft in vitro[38]. EPC-seeded grafts remained patent for 130 days as a carotid interposition graft in sheep. The EPC-explanted grafts exhibited contractile activity and nitric oxide-mediated vascular relaxation that were similar to native carotid arteries. These results indicate that EPCs can function similarly to arterial ECs and they could be a good EC source for vessel engineering.

Regarding SMCs, studies have shown that bone marrow derived mesenchymal stem cells (BMSCs) could be differentiated into SMC phenotypic cells in the presence of certain factors [40–42]. By using bone marrow derived ECs and SMCs from canine, Cho et al. have successfully engineered small-diameter vascular grafts in vitro[43]. The grafts remained patent for up to 8 weeks in the canine carotid artery interposition model. Cells labelled with a fluorescent dye prior to implantation were detected in the retrieved vascular grafts, indicating that the BMSCs participated in the vascular tissue regeneration. This work was confirmed by Koike et al. that networks of long-lasting blood vessels were formed in mice by co-implantation of vascular ECs and mesenchymal precursor cells. The networks were stable and functional for one year in vivo[44]. Moreover, Shin'oka and Mastumura et al. have successfully repaired the vessel defects with bone marrow derived cells in canine model as well as in patients [45–48]. These studies demonstrate that BMSCs could be a good SMC source for vessel engineering.

Adipose tissue is another stem cell source for ECs and SMCs. In 1983, Kern et al. isolated the microvascular ECs from human adipose tissue. These cells could grow readily to confluence and survived serial passages [49]. Arts et al. reported that microvasuclar ECs could be enriched from human adipose tissue by CD34 expression [50]. Martinez-Estrada et al. isolated an endothelial progenitor cell population that expresses Flk-1 from adipose tissue by three-dimensional culture. These cells could differentiate into mature ECs [51]. Meanwhile, Zuk et al. isolated another multi-potent population termed ‘adipose derived stromal cells (ADSCs)’ from adipose tissue, which can differentiate into adipocyte, osteoblasts and muscle cells [52]. Our recent work found that ADSCs isolated from human lipoaspirate could be induced to differentiate into SMCs. In addition, an elastic vessel wall could be successfully constructed in a bioreactor by seeding those cells on polyglycolic acid (PGA) scaffold (unpublished data). However, it should be noted that both BMSCs and ADSCs are multi-potent cells, whether cells would differentiate into other cell types (osteoblasts or adipocytes) and raise pathological problems after transplantation is unclear. Long-term fate of the cells after transplantation needs to be followed up in animal models.

Other cell types

Besides the stem cells, studies have also tried to replace the ECs and SMCs by other type of cells. L'Heureux et al. used adult human fibroblasts extracted from skin biopsies to construct TEBVs, which were further served as arterial bypass grafts in long-term animal models [53]. The TEBVs were antithrombogenic and mechanically stable for 8 months in vivo. Histological analysis showed a smooth muscle-specific α-actin positive cell population developed within the TEBV, indicating a complete re-generation of a vascular media. Campbell et al. took another approach of implanting silastic tubing into the peritoneal cavities of rabbits or rats to generate tissue tubes by an inflammatory response [54]. The tissue tubes that contained layers of myofibroblasts covered by a single layer of mesothelial cells could replace the aorta in rat. A patency rate of 68% was achieved in the absence of any heparin or spasmolytics over a period of up to 4 months. In addition, vascular re-modelling was observed 3 months after transplantation. Similar results have been achieved in other animal model [55]. However, no study has been carried out in human beings.

Biodegradable scaffolds

Scaffold is another key factor for tissue engineering. The 3-dimentional structure of scaffold provide a template for supporting cell growth, migration, differentiation and secretion of extracellular matrix (ECM) proteins, as well as for directing new tissue formation in the tissue regeneration process. Ideally, the scaffolds will be slowly resorbed in culture or after implantation, leaving only the tissue generated by the cells. In order to engineer a biocompatible vessel with growth potential and to avoid material-related side effects, the ideal scaffold for vessel engineering should be biodegradable. Varieties of materials have been utilized for vessel engineering, including the nature proteins, synthetic biodegradable polymers and decellularized vessels. The progresses of each material are discussed bellow.

Nature protein scaffolds

Nature proteins, such as collagen, elastin, fibronectin are the major components of ECM in the body. They are the most ideal substrates for cell attachment and cell signalling. Collagen and elastin are also the major components of blood vessel wall. Collagen gel has been used to create the first tissue-engineered vascular graft by Weinberg and Bell [7]. However, due to the inherent physical weakness of collagen gels and limited extracellular deposition by cultured SMCs, the mechanics of the grafts were not strong enough to support the physical load imposed by the haemodynamic environment. Many strategies have been tried to improve the strength of the collagen gel-based grafts, including the use of glycation to stiffen and strengthen collagen gel construct [56], the use of un-degradable or degradable meshes as ‘sleeves’[57–60], as well as the application of dynamic mechanical stimulation [61, 62]. Wrapping the constructs with Dacron mesh or polyurethane film could improve the strength of the grafts [57, 60]. This was further improved by wrapping with biodegradable materials, such as cross-linked type-I collagen and elastin [58, 59]. Furthermore, Boland et al. have applied the electrospinning technology to develop the biomimetic vascular constructs of micro-and nano-fibrous scaffolds from collagen and elastin, which could withstand the high pressure and pulsatile environment of the bloodstream [63]. As reviewed by Patel et al., elastin is a critical structural and regulatory matrix protein and plays an important and dominant role by conferring elasticity to the vessel wall [64]. Long and Tranquillo found that SMCs secreted more elastin on fibrin gels than on collagen gels [65]. Elastic small-diameter blood vessels were successfully engineered by using fibrin gels as scaffolds [66–69]. The grafts kept patent in jugular veins of lambs up to 15 weeks of observation [69]. Implanted vessels gained significant mechanical strength and reactivity that were comparable to those of native veins, indicating that fibrin-based TEBVs hold significant promise for treatment of vascular disease.

Biodegradable polymer scaffolds

Comparing with nature proteins, sythetic polymers are easily available and cheap. There is little or no batch-to-batch variations. In addition, polymers could be precisely modified to adjust their degradation rate, biocompatibility, elasticity. Several biodegradable synthetic polymer scaffolds have been investigated for their suitability in vascular engineering. Polyglycolic acid (PGA) is one of the most commonly used. By using PGA scaffolds and a biomimetic perfusion system, Niklason et al. produced the first autologous vascular graft and implanted into the arterial system [14]. The grafts were patent in vivo up to 1 month of observation. Although PGA fibre has good biocompatibility, its breakdown products are acidic, which could induce inflammatory response. Higgins et al. found that PGA breakdown products could lead to the dedifferentiation and decreased mitosis in SMC [70]. Moreover, PGA degraded too fast that result in a low mechanical property of engineered graft [14]. Other synthetic polymers with slow degradation rate, such as poly (L-lactic acid) (PLLA) [71], co-polymer of poly(D,L-lactic-co-glycolic acid) (PLGA) [71], poly 4-hydroxybutyrate (P4HB) [72] and co-polymer of PGA and polyhydroxyalkanoate (PHA) [73], have also been testified in vessel engineering. The biocompatibilities of the scaffolds were further improved by physical or chemical surface modifications [74–76]. However, the cellular toxicity of the breakdown products of those materials should be investigated in long-term study.

Decellularized vessels

Decellularized vessels, which are entirely composed of natural ECM, have good biocompatibility and could mostly maintain the mechanical properties of nature vessels [77]. Decellularization is typically accomplished by treating tissues with a combination of detergents, enzyme inhibitors and buffers. Although decellularized porcine carotid arteries followed by heparinization were successfully used to repair the abdominal artery without seeding cells in dog model [78], this approach is more difficult with human beings due to the lack of antithrombogenic EC layer on the lumen of the grafts. Teebken et al. obtained vessel grafts with stable biomechanical properties by seeding ECs and myofibroblasts from human saphenous veins on decellularized porcine aortas [79]. Similar work was performed by seeding human umbilical vein ECs or adult human vascular SMCs onto the decellularized porcine aortas after different decellularization processes [80, 81]. However, studies found that cell migration into these scaffolds was inadequate due to the very tight matrix organization specific to the aortic structure. To address this problem, Simionescu et al. prepared pure elastin scaffolds and pure collagen scaffolds by selectively removing the collagen component or elastin to create more porous scaffolds for cell infiltration [82]. Enhanced potential for repopulation by host cells in vivo was observed after subdermal implantation. In addition, new collagen fibres and bundles were found within the re-modelled elastin scaffolds and new elastin fibres within collagen scaffolds, respectively, indicating that they are able to support de novo ECM synthesis.

Porcine arteries are easy to access. However, the risk of transmission of animal pathogens to human being is still a big concern, even though it has been reported that decellularized porcine vascular scaffolds did not cause cross-species transmission of porcine endogenous retrovirus in a sheep model [83]. To avoid such problem, vessels from human being are the optimal choice [84, 85]. Daniel et al. decellularized HUV using an automated dissection methodology and created a promising scaffold that has excellent potential for cellular integration and maintain the mechanical properties of the native blood vessels [85]. The HUV scaffold could be a good candidate for vascular engineering.

Other materials

Hybrid materials of combining nature proteins and synthetic polymers have been testified for vessel engineering. Li et al. fabricated the vascular graft scaffolds using co-electrospun of PLGA, gelatin and α-elastin [86], while Stitzel et al. modified the approach by co-electrospun of PLGA, type I collagen and elastin [87]. No local or systemic toxic effects were observed when implanted the scaffolds in vivo. The scaffolds possessed tissue composition and mechanical properties similar to native vessels. The electrospun vessel matrix with both nature and synthetic materials could serve as a good scaffold for functional vessel engineering.

Vessel engineering in vitro

Vessel-reactors

Due to the dynamic environment of the cardiovascular system, the engineered vessel should be fully functional at the time of transplantation, which should be non-thrombogenic and have good mechanical strength and vasoreactivity. In addition, the mechanical and haemodynamic properties of vessel grafts are also crucial for their long-term survival [88]. To achieve such a functional graft in culture, a construct technology that could mimic the physiological vessel environment is required. Vessel-reactors have been developed and successfully utilized for vessel engineering by many groups [13, 14, 53, 89–91]. Basically, the vessel-reactor mimics the physiological stimuli that a native vessel received in the body, including the cyclic strain and shear stress [92–97]. Cyclic strain could significantly improve the mechanical property of engineered vessel, while shear stress could change cell alignment and improve endothelial cell adhesion [98–100]. Moore et al. demonstrated that cyclic strain could reduce the cell death [101], while Seliktar et al. found that cyclic strain could increase the matrix re-modelling by overexpression matrix re-modelling enzymes, such as matrix metal-loproteinase 2 (MMP-2) [102, 103]. In addition, Nikolovski et al. reported that cyclic strain could inhibit the switching of SMCs to an osteoblast-like phenotype in culture [104], which may prevent the un-wanted calcification in vascular graft. The important role of mechanics in vascular tissue engineering has been well reviewed by Nerem [100]. Meanwhile, it should be noted that optimal loading of stress to the graft is also critical. Liu reviewed that increased tensile stress and strain may induce vascular hypertrophy, and initiate focal atherosclerosis and intimal hyperplasia [105]. Solan et al. compared different rates of radial distension in vessel engineering, the adult heart rate (90 bpm) and the foetal heart rate (165 bpm) [106]. After 7 weeks of dynamic culture, no significant differences were observed between those two groups in terms of collagen and metallo-proteinase type 1 (MMP-1) expression. The parameters of vessel-reactor, such as pulse rate, deformation rate and pulsatile pressure, need to be optimized in future studies.

Culture additives

Besides the mechanical stimulation, chemical reagents and growth factors in culture media could also regulate the mechanic property of engineered graft. Transforming growth factor β1 (TGF- β1) could dramatically increase ECM production and deposition in plate culture [107]. Combination of TGF- β1, insulin and aprotinin could result in a significant improvement of both mechanical strength and vasoreactivity [68]. Ogle et al. reported that retinoic acid and ascorbic acid treatment could significantly elevate collagen and elastin gene expression, result in more ECM composition and enhanced mechanical properties of engineered graft [108]. Joddar et al. found that fragmented hyaluronan (HA) could stimulate the cell proliferation and synthesis of matrix elastin of SMCs in plate culture [109]. All these studies indicate that proper supplementation in the culture media is also helpful for vessel engineering in vitro.

Clinical applications

To date, TEBVs could be successfully constructed in vitro, and be used to repair the vascular defects in animal models [13, 43, 45, 47, 53, 69, 72, 89, 110]. However, only a few have achieved clinical success with this approach. The first clinical application of using an engineered vessel based on biodegradable scaffold was reported by Shin'oka et al. in Tokyo Women's Medical University [111]. The peripheral pulmonary artery was successfully reconstructed in a 4-year-old girl with the patient's own venous cells seeded onto a polycaprolactone–polylactic acid copolymer tube that was reinforced with woven PGA. After that, three patients were treated with same approach. In their following study, they took another approach of using autologous bone marrow cells as a cell source to avoid the time-consuming cell culturing step [46, 48]. Cells were harvested on the day of surgery, seeded directly on the polymer tube and the grafts were implanted right after 2–4 hrs of in vitro incubation. Twenty-three tissue-engineered conduits and 19 tissue-engineered patches were implanted for the repair of congenital heart defects [48]. They reported over 95% patency at 1 year without evidence of aneurysm formation or calcification. Moreover, there were no complications such as thrombosis, stenosis and obstruction of the tissue-engineered grafts. Long-term follow-up is desired to confirm the durability of this approach.

Different from the general approach, L'Heureux et al. successfully constructed vessel grafts using a novel method termed ‘Sheet-Based Tissue Engineering’[13]. In this approach, SMCs or fibroblasts were cultured in conditions that promote ECM deposition to produce a cohesive sheet that can be detached from the culture flask. The cell sheets were then rolled over a mandrel to form a vascular wall media without synthetic or exogenous scaffolds. After maturation, the inner tube was seeded with ECs. Using adult human fibroblasts extracted from skin biopsies, they successfully constructed TEBVs that could serve as arterial bypass grafts in long-term animal models [13]. This approach has been further testified in patients with haemodialysis [112]. The vessels constructed from autologous dermal fibroblasts and ECs were implanted as arteriovenous fistulas for dialysis access and were allowed to mature in vivo before use. During up to 5 months of implantation, no failures were observed with the first three patients, and the grafts were functioning well for haemodialysis access. These results are extremely encouraging. However, this approach is time consuming that would limit the application of these vessels in urgent cases.

Future perspectives

By seeding vascular cells on biodegradable scaffolds and further maturation of engineered vessel in bioreactor, successful results have been achieved in animal studies (Fig. 1). However, clinic trials demonstrated that the above elements and procedures are not indispensable for vessel regeneration; L'Heureux-engineered vessel grafts without exogenous scaffolds [112], while Shin'oka achieved vessel re-generation without in vitro culture [48]. From a surgeon's point of view, an off-the-shelf graft that available at any time for any patient is preferred. To achieve this goal, alternative ways beyond the traditional tissue engineering approach should be considered. As mentioned early, the basic requirements for a vessel graft should be non-thrombogenic, and have good mechanical strength and vasoreactivity. Those characteristics are accomplished by ECs and SMCs respectively in native vessel. If the scaffold alone could meet both of the characteristics, the cell seeding procedure could be possibly avoided.

Studies have demonstrated that native SMCs could migrate into the scaffold, and participate in the new vessel regeneration when scaffold alone were implanted [78, 110]. Theoretically, if the mechanical property of scaffold alone is close enough to the native vessel at the time of transplantation, the early involvement of SMCs in the graft is not necessary. As far as the vascular media could be completely regenerated by native SMCs before scaffold degradation in the body, seeding of exogenous SMCs could be avoided. Current technologies in material science are feasible to create such a scaffold by delicate design. In addition, certain modifications should also be considered to promote the SMCs migration, such as increasing scaffold porosity or embedding growth factors in the scaffold. Different from SMCs, early involvement of ECs in the graft is extremely important for preventing thrombosis at the time of implantation, which can not be easily achieved by scaffold alone at this moment. Thus, in a short-term plan, engineering vessel graft by seeding of ECs on a proper biodegradable scaffold might be the best way to promote the use of TEBVs in clinical application. In a long-term plan, scaffold that possesses antithrom-bogenic capacity at early stage and could recruit ECs at late time points is expected. Other strategies, such as delivery of therapeutic genes, have become of particular interests in tissue engineering [113, 114].

No matter which approach is taken, from a translational point of view, pre-clinical studies of human TEBVs need to be carried out in immunocompetent large animal models, and long-term outcomes should be followed up. For such experiments, development of an immunologically humanized animal is crucial. Recently, Zeng et al. generated a goat model carrying human cells by transplanting human cord blood cells into foetal goats at 45–55 days of gestation [115]. Long-term engraftment of human cells was detected in haematopoietic and non-haematopoietic organs for up to 2 years. Theoretically, the goat should have been immunologically humanized that would not reject human cells. Detail studies are going on to confirm the speculation. Hopefully, this xeno-transplant goat could provide a unique model for the evaluation of engineered human tissue graft in vivo.