*3.3. Biopolymers*

Besides the use of non-degradable and degradable vascular grafts, conduits based on natural matrices and proteins have also been proposed as an alternative solution (Table 3) [1]. These proteins can be used as the structural basis for the development of SDVGs, providing an appealing 3D microenvironment with proper binding sites for the cellular populations [47,48]. Several methods have been proposed to properly produce biopolymer-based SDVGs, including electrospinning, freeze-drying, and mold casting.

Collagen and its isoforms are the most abundant proteins that can be easily isolated, manipulated, and used for scaffold production, including also the engineered SDVGs. Habermehl et al. [89] optimized the procedure for collagen isolation from rat tails, and since then, a wide number of applications, where this structural protein is the main player, have been reported [90–93]. Until now, 28 different collagen types have been reported [94]. The collagen structure is composed of a repeated triple helix of proline (X) and hydroxyproline (Y). Based on the triple helix organization, collagen can be distinguished into a) fibrils (including types I-V and XI), networks (including types IV-X), and filaments (including type VI) [94]. Among them, collagen I is the most abundant type in mammalians, composed of two α1(I) and one α2(I) chains. Collagen type I offers a grea<sup>t</sup> number of integrin-binding sites, which can control the cell adhesion, differentiation, and overall cellular behavior. Different types of collagen-based scaffolds have been used in tissue engineering applications [91]. Collagen scaffolds combined with hydroxyapatite have been used in orthopedic applications, inducing bone and cartilage regeneration [95,96]. Moreover, collagen has been proposed as a drug delivery system (DDS) to release pro-angiogenic factors for wound healing applications and as a natural coating of vascular grafts [97,98]. However, collagen is characterized by low mechanical properties and increased thrombogenicity [99,100]. For this purpose, cross-linking with fixative agents such as glutaldeyhyde has been proposed [101,102]. Nevertheless, the improvement in mechanical properties, severe cytotoxicity are accompanied most of the time due to the crosslinking agen<sup>t</sup> that was applied [103]. Alternative crosslinking methods have also been utilized such as photo-crosslinking or the use of carbodiimide [104–106]. Moreover, collagen-based SDVGs combined with fibronectin or elastin fibers have shown promising results regarding the biomechanical and anti-thrombogenic properties. Another promising biomaterial for SDVGs fabrication is the silk fibroin [107,108]. Fibroin is derived from Bombyx mori (silkworm) and is composed of β-sheet crystal and semicrystalline regions occurred after the removal of sericin [109,110]. Sericin is a highly antigenic protein, which covers the silk fibers [111]. Additionally, fibroin has anti-thrombogenic properties and can be degraded over time, therefore, could be a valuable source for the production of SDVGs [112]. Enomoto et al. [113] managed to develop a fibroin-based SDVG whose patency was compared with ePTFE vessel conduits. In this study, the developed conduits (d = 1.5 mm, l = 10 mm) were implanted as abdominal aorta interposition grafts in male Sprague-Dawley rats for a time period of 72 weeks [113]. Fibroin based SDVGs remained patent (85%) over 64 weeks, whereas ePTFE grafts were patent (48%) for 32 weeks [113]. In addition, an increased number of SMCs and ECs was observed in fibroin-based SDVGs compared to ePTFE grafts, reflecting in this way the impaired overall functionality of the latter.

To date, fibrin, which can be obtained from peripheral blood, comprises a biomaterial that can be applied in SDVG engineering [114,115]. Fibrin is produced through the cleavage of fibrinogen [115]. Fibrinogen (MW: 340 kDa), a glycoprotein that is abundant in plasma, contains three pairs of polypeptide chains, the A α, Bβ, and γ, which are connected with 29 disulfide bonds [115]. Upon stimulation, thrombin cleaves the fibrinopeptides A and B, between Arg-Gly residues [115,116]. The remained (<sup>α</sup>, β, γ)2 can be polymerized with other fibrin molecules, resulting in the production of fibrin final form. Due to that, fibrin can be produced from patients' blood, without causing any negative adverse reactions to the recipient [115,116]. Fibrin is a rich source of growth factors, cytokines, and chemokines, such as Tumor Necrosis Factor-A (TNF-A), Vascular Endothelial Growth Factor (VEGF), Fibroblast Growth Factor (FGF), Platelet-Derived Growth Factor AA (PDGF-AA), Interleukin 1A (IL-1A), IL-1B, IL-2, IL-6, IL-8, TNF-Receptor type-1 associated Death domain protein (TRADD), CC-motif chemokine receptor 1, etc. [117,118]. Recently, platelet-rich plasma or fibrin gel have been employed in a series of regenerative medicine applications such as skin wound healing and dystrophic recessive epidermolysis bullosa [119,120]. Except for the patient's own blood, fibrin can be produced from other sources like the umbilical cord blood (UCB). Rebulla et al. [121] initially optimized the PRP and fibrin production from UCB units that did not meet the criteria for cryopreservation. In addition, our group suggested a protocol for the e fficient production of PRP and fibrin from low volume CBU units [118]. The development of allogeneic fibrin holds significant advantages such as the avoidance of repeating blood sampling, especially from severe conditioned or elderly individuals, low immunogenicity of the obtained fibrin, and absence of allergic reaction [118].

Recently, fibrin has been employed in vascular tissue engineering. In the beginning, fibrin was used as a coating in collagen-based vascular grafts [122]. To date, the research society is performing a significant e ffort to produce fibrin-based vascular grafts [123,124]. Most of the time, a pulsatile bioreactor system is required for the proper maturation of the developed vascular grafts. Moreover, approaches, where ECs and SMCs are utilized in fibrin-based vascular grafts, have been proposed [36]. Swartz et al. [125] used recellularized fibrin-based vascular grafts as implants in a sheep model. Specifically, these grafts were implanted in the jugular veins for a time period of 15 weeks. Histological analysis of the grafts showed the successful in vivo remodeling, where collagen and elastin depositions were evident [125]. However, the fibrin-based vascular grafts were characterized by impaired biomechanical properties. Indeed, the average burst pressure of fibrin-based vascular grafts was 543 ± 77 mmHg, which is very low to withstand the physiological burst pressures of blood flow [125]. A recent study from Yang et al. [126] showed that the mechanical properties of these vessel conduits can be improved with the addition of PCL, resulting in the production of a hybrid graft (fibrin-PCL vascular graft). In this study, electrospun PCL/fibrin vascular grafts were developed, followed by evaluation of mechanical properties, cytotoxic e ffects, and in vivo biocompatibility [126]. The burst pressure of these hybrid vascular grafts was 1811 ± 101 mmHg, which is similar to native

blood vessels (2000 mmHg). Furthermore, no cytotoxic effects or in vivo immune response were reported, in this study [126].

The production of vascular grafts made of chitosan has also been reported [127]. Chitosan is a linear polysaccharide that is closely related to sulfated glycosaminoglycans (sGAGs) [128]. Chitosan is a natural material that is derived from the shell of shrimps and crabs and has been used extensively in a wide range of tissue engineering applications [128]. Specifically, chitosan has been used for the production of hydrogels, DDS, coatings, and also in wound healing applications [128]. In addition, chitosan can be combined with degradable polymers such as PCL and PLA for scaffold fabrication [129]. Moreover, chitosan has mild antibacterial properties, which are beneficial for in vivo applications [130]. Recently, the use of chitosan has been proposed for the development of SDVGs. In the context of vascular graft production, the electrospinning technology can be utilized to produce conduits with wide pore distribution, high porosity, and adequate microenvironment for cell adhesion and proliferation. Wang et al. [127] reported the development of a PCL/chitosan (PCL/Ch) hybrid-based SDVG with anti-thrombogenic and anti-bacterial properties. For the scaffold fabrication, the electrospinning technology was utilized [127]. The results of this study showed that the PCL/Ch hybrid-based SDVGs have similar anti-thrombogenic properties as the heparin-coated vessel conduits, while the bacterial killing ratios were 64% for *S. aureus* and 73% *E. coli* [127]. Yao et al. [129] also developed electrospun PCL/Ch SDVGs, which were further combined with heparin and referred as Hep-PC/Ch grafts. These grafts were further implanted as aortic replacements in male Sprague-Dawley rats. Their functionality was compared with PCL/Ch vascular grafts (without heparin immobilization). After 4 weeks of implantation in rats, the PCL/Ch explants were characterized by thrombus formation, while no thrombus formation was observed to Hep-PCL/Ch grafts [129]. Furthermore, Hep-PCL/Ch grafts were characterized by good patency rate and successful endothelialization as were indicated by SEM analysis [129]. Taking into consideration the above data, chitosan is a material that can be used in combination with degradable polymers to produce functional SDVGs.


**Table 3.** Representative applications of TEVGs derived from biopolymers.


**Table 3.** *Cont.*

## *3.4. Hybrid Polymers*

The proper combination of synthetic and natural polymers could produce functional engineered SDVGs. These conduits combine the beneficial features of both materials and are characterized by improved biomechanical, anti-thrombogenic, anti-bacterial, and cell adhesion properties [139]. Furthermore, hybrid vascular grafts can be combined with key specific growth factors such as TGF-β1, VEGF, EGF, HGF, etc., which can be accumulated in the vascular wall [129,140,141]. These growth factors can be spatially released from there, affecting in this way several cellular functions including cell migration and growth [142]. To date, there is an increasing number of research teams, which are focusing on the production of hybrid vessel conduits (Table 4). Tillman et al. [143] produced a PCL/collagen vascular graft, with improved biomechanical properties. The PCL/collagen conduits served as the aorta and iliac artery interposition grafts in rabbits and remained for a time period of 1 month [143]. These hybrid grafts were free of any aneurism or thrombus formation, while Doppler ultrasound showed good patency (85%) of the grafts. Histological analysis of the explants revealed the absence of inflammation, thus completely lacking any infiltrating immune cell [143]. Wise et al. [144] produced PCL/elastin vascular grafts, where parameters such as ECs adhesion and proliferation, blood biocompatibility, burst pressure, and in vivo functionality were assessed [144]. Specifically, these grafts were able to be recellularize both in vitro and in vivo with the ECs. The burst pressure of the grafts was 1500 ± 150 mmHg; however, it was less than the minimum burst pressure that was evident in human native blood vessels (1700 mmHg). Similar good patency and cell infiltration results of hybrid acellular vascular grafts have been reported in the literature [144–147].

In addition to these fabrication strategies, the use of cellularized hybrid vascular grafts may provide better outcomes regarding the mechanical properties and overall patency [34]. Thomas and Nair [148] developed a vascular graft, which was composed of gelatin/vinyl acetate copolymers, utilizing the electrospinning method. The composed vascular grafts were successfully recellularized with murine SMCs, followed by maturation with a pulsatile bioreactor system [148]. The pulsatile forces, generated by the bioreactor, effectively stimulated the SMCs migration, proliferation, and gene and protein expression [149].


**Table 4.** Representative applications of TEVGs derived from hybrid materials.

The overwhelming increase of new CVD cases each year is leading to the exploration of alternative sources for the production of engineered SDVGs. Most of these approaches, including non-degradable, degradable, and biopolymer grafts, are requiring extended evaluation, while their proper fabrication could last over 28 days [2,34]. Toward these shortcomings, the hybrid-based TEVGs may pose a reliable approach, reducing the manufacturing time and thus producing SDVGs with improved properties. However, more research is needed to be performed in order for the hybrid SDVGs to be readily used by clinicians in cardiovascular reconstructive surgery.

#### **4. Decellularized Vascular Grafts**

In the last decade, the application of the decellularization method for the production of vascular grafts has gained significant attention from the scientific society [154]. Decellularization aims to remove completely the cellular material from the tissue while preserving the ultrastructure of ECM. Depending on the tissue source, different decellularization approaches may be applied to produce effectively an acellular matrix [154,155]. Until now, decellularization has been applied successfully to a grea<sup>t</sup> number of organs and tissues, including lung, liver, kidney, heart, cartilage, etc.

#### *4.1. Decellularization as a Method for the Production of Vascular Grafts*

To achieve the production of an acellular matrix, different decellularization protocols may be used. Mostly, the decellularization protocols include physical, chemical, enzymatic, or a combination of those methods to acquire the best outcome [154–156]. The initial step of the decellularization approach is cell destruction through the solubilization of the cytoplasmic membrane and DNA fragmentation. Then, the cellular and nuclear debris must be completely removed from the tissue's ECM. Excessive removal of decellularization solutions also is an important step of the process to limit the possibility of any cytotoxic effects [154–156]. The final step of the decellularization procedure is the sterilization of the produced scaffold.

Sterilization can be achieved either by immersion of the scaffold into antibiotic solutions or by applying physical methods such as UV and γ-irradiation [155,156].

The increased global demand for vascular grafts led the researchers to evaluate further the decellularization approach for the production of vessel conduits [36]. Large- and small-diameter vascular grafts have been decellularized with the application of different decellularization approaches. Mostly, a proper combination of the decellularization approaches, such as snap freezing, use of ionic and non-ionic detergents, trypsin addition, and mechanical agitation or sonication, have been utilized [154,156]. Among them, the use of chemical compounds in combination with physical methods, is the most effective and safe for producing acellular vascular grafts. The most used detergents for the decellularization procedure are sodium dodecyl sulfate (SDS), sodium deoxycholate (SD), Triton X-100, Triton X-200, 3-[(3-Cholamidopropyl)dimethylammonio]-1- propane sulfonate (CHAPS), and Ethylenediaminetetraacetic acid (EDTA) [156]. Additionally, in the literature, the combination of hypotonic and hypertonic treatments, with enzymatic digestion, has been also reported for the efficient production of acellular SDVGs [154–156].

Taking into consideration the above data, vascular grafts and especially SDVGs can be derived from various sources such as animals (porcine or sheep) or cadaver donors, effectively decellularized, and immediately used (Figure 2). However, significant drawbacks are accompanying the above proposal. In the past, a grea<sup>t</sup> effort regarding the use of animal-derived TEVGs in human applications was performed [157]. Despite the complete removal of cellular and nuclear materials from the vessel's ECM, animal-derived vascular grafts can induce an extended immune response due to the presence of alpha-gal-epitope (Galalpha1-3Galbeta1-(3)4GlcNAc-R) [158]. This epitope is abundant in non-primates and New World monkeys and synthesized by the alpha1,3galactosyltransferase (alpha1,3 GT) [158–160]. On the other hand, humans, apes, and Old World monkeys produce anti-Gal antibodies, which are representing 1% of the circulating immunoglobulins [158–160]. In this way, human recipients when receiving animal-derived vascular grafts, exert significant immune response against the aforementioned grafts, which finally leads to graft rejection. Nowadays, much effort has been focused on the cleavage of a-gal epitope or the production of transgenic animals (without the presence of a-gal epitope), although more research must be performed toward this direction [161,162]. Recently, genome editing with CRISPR-Cas9 may assist in this field [163]. Cadaver donors may constitute an alternative source, for obtaining SDVGs. Based on organ donation statistics, only 3 in 1000 people find suitable organs

and more than 112,000 people are waiting for organ transplantation [164]. Furthermore, the bioethics rules must be modified in order to allow organ transplantation and especially vessel transplantation. The production of vascular grafts with the decellularization approach may be a promising approach, thus increasing the number of available transplantable vessels.

**Figure 2.** Production of SDVGs with the decellularization approach. Initially, vascular grafts can be obtained either from human cadaveric or animal donor. Then, the obtained SDVG is submitted to decellularization to remove the residual cellular population. The produced acellular vascular graft can be either implanted back to the patient or to submitted in recellularization with the patient's own cells. The cellular populations can be isolated and expanded from the patient's tissue biopsy. When cells reached the desired cell number, they can be used for the recellularization of the acellular vascular graft. Finally, the produced cellularized SDVG can be implanted in the patient. The whole procedure can be performed under good manufacturing practice (GMP) conditions.

#### *4.2. Establishment of the Decellularization Approach*

The production of a completely acellular scaffold is a demanding process; however, most of the time a small quantity of residual cellular and nuclear materials are evident. Different decellularization methods are characterized by variable results, indicating that the majority of them cannot produce a completely acellular scaffold. The presence of the cellular components could induce an immune response and hyperacute reaction by the host upon implantation [165,166]. This could lead to unfavorable adverse reactions, resulting in graft occlusion, calcification, and rejection, with the majority of them to be life-threatening for the recipients. Globally, several researchers have tried to validate the decellularization approach in different tissues and organs, leading them to several criteria for the establishment of the successful decellularization approach. Among them, Gilbert et al., Crapo et al., and Badylak et al. have performed the most valuable work, proposing the following criteria [154,156,167].


• Lack of visible nuclear materials, either with <sup>4</sup>,6-diamidino-2-Phenylindole (DAPI) or hematoxylin and eosin (H&E)

Except for the above-mentioned criteria, the total amount of DNA including single-stranded (ss) and ds, should also be quantified and taken into account. DNA quantification can be performed photometrically, or with the use of different commercial kits such as the Picogreen Assay. Indeed, there are numerous studies where Picogreen assay is the optimum method for the quantification of the DNA in decellularized matrices [168–171]. However, the PicoGreen assay can detect only the ds DNA, while the ss DNA cannot be quantified. On the other hand, the spectrophotometric quantification of DNA by measuring the ratio of absorbance 260 nm/280 nm, may provide more data regarding the presence of the total DNA in the acellular scaffold [172]. It is known that either ss or ds DNA can induce the host's immune reaction, and the accurate DNA quantification is of major importance. Furthermore, the remaining DNA in the scaffolds can be evaluated through the performance of gel electrophoresis [172]. Typically, the DNA samples can be loaded onto 1–2% *w*/*v* agarose gels and observed under UV light. The absence of dense DNA bands or bands with less than 200 bp DNA confirms further the successful decellularization [172].

The last criterion involves the observation of the tissue sections for any possible nuclear material either with H&E or DAPI [154,156]. H&E is the first-line histological stain that is performed in order to properly evaluate the success of the decellularization approach. The absence of black stain in the histological samples indicates the loss of nuclear material. Besides H&E, more specific stains can be applied for the determination of decellularization. Masson's trichrome (MT), which stains collagen (blue), muscle cells (red), and nuclear materials (black), can be used for the evaluation of the presence or absence of SMCs. Except for the content of the cellular population, this stain can indicate the proper preservation of the collagen fibers in the acellular scaffold. In the same way, Elastic van Gieson (EVG) can stain simultaneously the elastic/collagen fibers and nuclear material [154,156].

Nevertheless, the production of a completely acellular scaffold is optimum, and the preservation of key ECM features such as the orientation of collagen and elastin fibers are also important. The microarchitecture structure of tissues and organs can determine the decellularization approach, which will be selected. Complex tissues, where the orientation of collagen and elastin determine eventually their biomechanical properties, can be decellularized with the use of non-enzymatic approaches [173–175]. It has been shown that trypsin can damage significantly the collagen fibers of a tissue, affecting in this way possibly the graft's biomechanical properties and cell-binding sites. A balance between the proper elimination of cellular components in combination with the minimum effect in ECM key proteins must be found when a decellularization protocol is applied.

#### *4.3. Decellularized Animal-Derived SDVGs*

The first attempt for establishing a decellularization protocol was performed in 1966, several years before the attempts of Weinberg and Bell for the manufacturing of synthetic polymer vascular graft [33]. Rosenberg et al. [176] applied for the first time an enzymatic decellularization protocol in bovine carotid arteries. The produced acellular vascular grafts were implanted in 16 patients as femoropopliteal and iliofemoral bypass grafts. The implanted grafts withstood the blood flow pressure; however, graft occlusion was reported during a time period of 2 years postoperatively [176]. Since then, new decellularization approaches have been found and validated in a wide range of tissues and organs including the vessels such as the aorta, carotid, and coronary arteries (Table 5) [177–180]. The first decellularized vascular grafts were derived from bovine vessels and ureters, which further became commercially available as Artegraft®, Solcograft®, ProCol® (LeMaitre Vascular, Inc., Burligton, MA, USA), etc. [2,181]. Today, several companies are focused on the production of decellularized vascular conduits based on the bovine vessels. However, the presence of a-gal is a significant limitation, and in order to overcome this issue, crosslinking with fixative agents such as glutaraldehyde is performed [103]. A significant drawback to this approach is the cytotoxicity mediated by the fixative agents, resulting in minimum applicability of those grafts [103]. However, modern fixative agents such

as carbodiimide with low or no cytotoxicity have been applied [106]. Another significant drawback of the crosslinked decellularized bovine blood vessels is the lack of in vivo remodeling properties, which makes them unavailable for applications in pediatric patients [182]. Additionally, it has been reported that decellularized animal-derived blood vessels are characterized by similar patency rates as synthetic vascular grafts [36].

Recently, the use of small intestine submucosa (SIS) has been also proposed for the production of large and small TEVGs [183]. Typically, SIS can be derived either from the porcine or ovine origin [184]. Decellularization can efficiently be applied in SIS, and then the produced material can be folded in a tubular mandrel to produce a vascular graft [184]. Moreover, crosslinking with fixative agents such as glutaraldehyde has been reported as an important step in the manufacturing process [185]. These grafts currently have been evaluated for their functionality in animal models, showing good patency rates [36]. Moreover, the patency rates were superior or equal to ePTFE grafts and native ovine artery [36].


**Table 5.** Summary of representative studies toward decellularized animal-derived vascular grafts.

Nowadays, the cost production of synthetic vascular grafts has been reduced and, considering the above data, their application is more preferable [36]. On the other hand, due to the increased demand for SDVGs, alternative sources must be explored in order to cardiovascular surgeons to have more available options.

#### *4.4. Decellularized Human-Derived SDVGs*

The first human blood vessel conduits served as transplants in reconstructive surgery were derived from human cadaver femoral veins. Indeed, human femoral arteries were submitted to decellularization to produce acellular vascular conduits [192]. These grafts were used initially as arteriovenous fistulas (AVF) allografts [193]. Furthermore, these grafts were commercialized under the name Synergraft® (CryoLife, Inc., Kennesaw, GA, USA) [34]. Decellularized iliac vein is another human vascular graft that has been proposed for vascular reconstruction applications [194]. Moreover, this graft was recellularized with the patient's cells such as ECs and SMCs and then was applied in a 10-year-old female patient with extrahepatic vein obstruction [195]. Before the cell seeding, the graft was evaluated for the presence of cell/nuclear materials and HLA class I and II genes. The operation was performed at Sahlgrenska University Hospital in Gothenburg, Sweden, and the outcomes were published in 2012 [195]. After 1 year of implantation, the graft was occluded, explanted, and a new vein graft was used. Finally, the patient responded well, no anti-endothelial cell antibodies were detected, and there was no need for receiving any immunosuppressive agents [195]. In this direction, human umbilical vessels may be an alternative source for the production of SDVGs [196]. The human umbilical cord (hUC) contains approximately two arteries and one vein, which are mediating in gas exchange and nutrient supply through the fetomaternal circulation [197]. The human umbilical arteries (hUAs) are responsible for the transportation of non-oxygenated blood from the fetus to the mother, while the human umbilical vein (hUV) performs exactly the opposite process [198]. The HUAs and hUVs are characterized by three layers, the inner (tunica intima), the media (tunica media), and the external layer (tunica adventitia). In addition, the hUAs and hUV are vessels without branches and their entire length can be varied and is dependent on hUCs length [197]. The length of a typical hUC is 20–60 cm and is characterized by an average number of 40 helical turns. In addition, hUAs are characterized by specific protrusions located in the tunica intima, throughout the entire vessels, which are known as "Hoboken valves" [199]. These valves prevent successfully the reflux of the non-oxygenated blood back to the fetus. Both vessels can easily and non-invasively be isolated from the hUC after gestation. Typically, in the case of using the human umbilical blood vessels, signed informed consent from the mothers must be obtained [196]. The informed consent should fulfill the requirements of the National law, regarding cord tissue donation and also should be in accordance with the Helsinki declaration.

HUV has been applied as a vascular bypass graft since 1974, followed by commercialization, which was known as Biograft® [200,201]. Several years later, the outcome of the use of Biograft® was evaluated. Specifically, a comprehensive evaluation of the use of hUV as a femoropopliteal bypass graft, a study involved 133 operations and a 5-year follow up, was performed [202]. In this study, it was shown that 6% of the patients died within 30 days after the implantation. The majority of the complications in patients were evident within the first 3 months postoperatively. The mean patency rate was 65% and 50% within the first and fifth year, respectively. No infection of the graft was reported in the current study [202]. The obtained results of the current study indicated that the stabilized hUV could potentially be used as a source for SDVG production. Currently, the gold standard autologous graft for coronary artery bypass surgeries is the SV [203]. However, other blood vessel sources have been evaluated such as the cephalic artery, stabilized hUV, and ePTFE grafts (Table 6). Among them, the hUV seems to share better patency and biocompatibility properties compared to the cephalic and ePTFE vessel conduits [204]. Indeed, a randomized clinical trial has shown that the patency rate of ePTFE was 40% within the first year of implantation, while stabilized hUV was 75% for the same time period [204]. Moreover, SV and stabilized hUV seems to share similar patency rates. Although the results were quite promising, the hUV was stopped to be used as a vessel substitute due to significant drawbacks [204]. HUV is more difficult to be applied technically than SV or synthetic conduits. Moreover, hUV may lack elasticity, making it more fragile [200–202,204,205]. In addition, the crosslinking reagents used for its stabilization like glutaraldehyde could induce severe cytotoxicity. Another drawback that is accompanied by the use of the crosslinking agents is the lack of in vivo remodeling properties, which make it less available for pediatric patients [204].

Taking into consideration the above data, the use of hUAs as possible vascular conduits should be also evaluated. Kerdjoudj et al. [206,207] used for the first time the human umbilical artery as potential small-diameter vascular grafts. Initially, this approach involved the deposition of a synthetic polyelectrolyte film in hUAs in order to avoid the platelet adhesion and eventually the thrombus formation [206,207]. In this study, the hUAs were enzymatically de-endothelialized and treated with poly(styrene sulfonate)/poly(allylamine hydrochloride) (PSS/PAH) to develop multilayers of polyelectrolyte film. This negative polyelectrolyte film can exert key anti-thrombogenic properties, avoiding in this way the platelet accumulation and thrombus formation in the lumen surface of the vessels. Then, these grafts were implanted as carotid interposition grafts in rabbits and remained for a time period of 3 months [206]. The results of this study were impressive, indicating the long-term patency (over 12 weeks) of the hUAs treated with PSS/PAH film. Furthermore, successful cell invasion of PECAM+ ECs and α-SMA<sup>+</sup> SMCs was evident in tunica intima and media, respectively [206]. Minimum intimal hyperplasia was reported in these grafts, which were mainly exerted through collagen production from SMCs [206]. Several months later, Gui et al. [208] evaluated a novel decellularization protocol in hUAs. In this study, a series of important experiments were performed, obtaining valuable information regarding the utilization of the hUAs as SDVGs [208]. Furthermore, the decellularized hUAs were implanted as abdominal interposition grafts in nude rats. After 8 weeks of implantation, thrombus formation was observed in the vascular grafts. Despite this drawback, decellularized hUAs sustained the blood flow and finally, the vessel did not rupture [208]. Several years later, the comprehensive proteomic analysis combined with histological data in native and decellularized hUAs was performed [209]. Until now, several researchers have evaluated the possibility of using the hUAs as transplants [206–211]. In 2020, our group showed that the decellularized hUA can be successfully vitrified and stored at −196 ◦C over a long time period [172]. Specifically, vitrified (decellularized) hUAs retained the ECM structure after 2 years of storage in liquid nitrogen. Furthermore, the vitrified grafts were used for common carotid bypass grafting in porcine animal models and remained for a time period of 1 month. Although the occurrence of platelet aggregation and thrombus formation was observed, vitrified hUAs were successfully in vivo remodeled [172]. The grafts were recellularized by the host's VSMCs, and due to the blood flow stress-strain forces, increased production of elastin fibers was occurred [172]. By the time that this publication is prepared, another work from our group is focused on the biomechanical and proteomic characteristics of the decellularized hUAs [212]. The proteomic results have been deposited to the ProteomeXchange Consortium with the dataset identifier PXD020187 (https://www.ebi.ac.uk/pride/) and are currently publicly available. In this study, a rapid decellularization protocol was effectively applied in hUAs. No cellular or nuclear remnants were evident, while at the same time the proteomic and biomechanical analysis showed the preservation of key ECM structural proteins and mechanical characteristics of the hUAs, respectively [212].

HUAs may represent a better source for the development of SDVGs compared to hUVs. However, extended validation experiments to better determine the stability and functionality of these grafts should be performed. The future goal will be the successful recellularization with ECs/VSMCs and implantation to large animal models for longer time periods to acquire more valuable data regarding the possible application of hUAs as SDVGs.


**Table 6.** Summary of representative studies toward decellularized human-derived vascular grafts.

#### *4.5. In Vivo Performance of Decellularized and Cellularized SDVGs*

Both decellularized and cellularized SDVGs have been tested in a wide series of experiments, including the evaluation of in vivo performance and biocompatibility [193,216]. It has been shown that decellularized SDVGs lack proper function and are characterized by a high probability of thrombus formation and graft failure [2]. Initially, the exposed collagen, located in the lumen side of the acellular SDVGs, triggers the platelets to aggregate [217]. The first step of this process involves the binding of the soluble form of von Willebrand factor (vWF) with the exposed collagen. Then, and upon vessel exposure to increased shear stress, the platelets are stimulated, leading to large aggregations development through the interaction between platelet glycoprotein (GP) Ib-V-IX receptor and vWF [218]. Furthermore, additional platelet receptors, including GPVI and <sup>α</sup>2β1, o ffer more stability to the developing thrombus [219]. Besides, the exposed collagen, fibronectin, and laminin assist in the development of thrombus, through the interaction with platelets' integrins <sup>α</sup>5β1 and <sup>α</sup>6β1, respectively [217]. Furthermore, VSMCs contribute significantly to vessel functions. VSMCs is a specific smooth muscle cell subset, located in the tunica media of the vessel wall, responsible for vasoconstriction and vasorelaxation [39]. In these processes, the role of ECs in the regulation of vascular tone is very important. Upon stimulation of ECs, due to high shear stress, the nitric oxide (NO) synthase is activated and increases the NO production, which can cause vasorelaxation. In addition, VSMCs produce high amounts of ECM proteins, including collagen, elastin, and fibronectin, contributing to the regeneration of the injured vessels. An additional important function of VSMCs is the ability to retain the circumferential orientation of the collagen and elastin fibers in the vessel wall [61]. It has been shown that the removal of VSMCs during the decellularization process could alter the alignment of the collagen and elastin fibers. The presence of uncrimped collagen and elastin fibers in decellularized vascular grafts can induce significant alteration to their biomechanical properties, resulting in mismatch compliance between resident and transplanted vessels [212].

On the other hand, cellularized engineered SDVGs are conduits with improved properties, which may result in a more favorable outcome upon implantation. For this purpose, vessel bioreactors such as pulsatile flow or dynamic culture systems, are currently used, which can result in vessel production with uniform coverage of ECs and VSMCs. It has been shown in the literature that recellularized TEVGs are characterized by a lower risk for thrombus formation and graft rejection compared to non-cellularized TEVGs. Zhou et al. [220] reported low neo-intimal formation in recellularized vessels, which were obtained from decellularized canine vessels. In addition, Kaushal et al. [221] and Ma et al. [222] showed that the presence of ECs and VSMCs in the manufactured TEVGs contributed to neo-intimal and neo-medial reconstitution. Row et al. [223] managed to trace the cells used for the recellularization of TEVGs postoperatively. The recellularized TEVGs were implanted as interposition grafts into the coronary artery in female sheep for a time period of three months. In this set of experiments, it was observed that donor cells (used for the recellularization) were gradually substituted by the host cells. Importantly, donor cells represented only 17% and 8% of the total cells after 1 and 8 months postoperatively. Furthermore, no T or B cells were evident in the implanted vessels, indicating that the recellularized vessel favors no immune response [223].

Considering the above data, recellularized TEVGs have greater possibilities to avoid neo-intima formation, thrombus development, and graft failure, and are superior to decellularized vascular grafts. Nevertheless, greater e ffort regarding the recellularization process must be performed by the research groups worldwide to produce properly functional TEVGs.

#### **5. Manufacturing Methods for the Development of SDVGs**

Globally, the increased demand for vascular grafts requires the production of readily available functional transplants, although their large-scale production is a quite challenging task. Currently, there is a wide variety of SDVG manufacturing methods, including tissue-engineering by self-assembly (TESA), electrospinning, and bioprinting [224–226]. Since the first attempts of Weinberg, Bell, and Rosenberg in manufacturing vessel conduits, vascular graft production has been evolved and

represents a quite interesting and interdisciplinary research field of the 21st century [33,176]. As has been proposed by Langer and Vaccanti, tissue engineering aims to the production of sca ffolds or matrices in order to replace or remove the damaged tissues [227]. In the same way, the development of vascular grafts is characterized by the same principles. A number of di fferent scientific areas must be combined to properly produce a vascular graft, which will replace the damaged one. Nowadays, the production and transplantation of LDVGs have been proven e fficient based on the performed clinical trials [1]. On the other hand, the production of fully functional SDVGs is still under the developmental stage, needing further exploration. A few companies have achieved significant progress in the production of readily available SDVGs. Among them, Cryolife, Artegraft, Integra, and Gore are providing ready to use vascular grafts, derived mostly from in-house production (synthetic grafts), animal origin (bovine and porcine vessels), and cadaveric donors (human vessels). However, the clinical utility of vessel transplants requires also the evaluation of alternative methods for the production of vascular grafts. The natural ECM structure and its key mechanical properties are di fficult to be reproduced with the aforementioned methods, but significant progress to this direction day by day is made.

#### *5.1. TESA Approach*

The TESA approach was developed by the pioneer L'Heureux and aimed at the production of vascular grafts utilizing cell sheets [224]. To achieve this outcome, no supporting vascular sca ffolds are required.

The basic concept has relied on the use of cell sheets containing fibroblasts, mesenchymal stromal cells (MSCs), ECs, and VSMCs, which were shaped around a mandrel to produce a tubular formation (Figure 3). Further maturation in the pulsatile bioreactor is required in order to vascular grafts to achieve the prerequisite burst pressure and overall mechanical properties [224,228]. The initial work of L' Heureux et al. [229] involved the cultivation of SMCs and fibroblasts in a standard culture medium contained sodium ascorbate. One month later, the produced sheets were shaped with the use of a tubular mandrel. The same technique was applied for the production of the di fferent layers of the vascular graft. The results of this study were impressive. Specifically, histological analysis revealed the proper localization of the cellular populations, while the burst pressure of the produced vascular conduit was more than 2500 mmHg [229]. Moreover, these vessels were implanted as femoral artery interposition grafts in canine animal models, withstood the blood flow, and met the fundamental requirements of a vascular graft. More experiments also were performed utilizing human cells for the development of vascular grafts and their testing in di fferent animal models. The above results led to the performance of the first clinical trial with TESA-produced TEVGs [229].

Between 2004 and 2007, a multicenter clinical trial was performed in patients who followed hemodialysis [230]. In this study, the vascular grafts were produced from the patient's own cells (fibroblasts and SMCs) utilizing the cell-sheet technique in the same way as previously has been described. The produced grafts were characterized by a mean burst pressure of 3512 mmHg and were used as AVF conduits. The patency rates of the grafts were 78% and 60% after the 1st and 6th months of implantation, respectively. The most observed complications involved the development of aneurism and lumen thrombosis. Despite these drawbacks, this study represents an initial step toward the clinical utility of TESA-produced vascular grafts [230].
