Extensive multidisciplinary research, especially in the latter half of the 20th century (early 1950s), has focused on the replacement of damaged, occluded, ruptured and atherosclerotic blood vessels [1
]. Candidate materials have evolved from inert gold tubular structures [2
] to the advanced synthetic prosthetic grafts found in modern clinical settings [1
]. Cardiovascular disease (CVD) is currently one of the leading causes of death in the Western world, hence vascular grafts with increasing requirements are being used to a greater degree and in a wider variety of clinical scenarios [1
The molecular mechanisms involved in prosthetic vascular graft healing, as well as detailed analysis of healing kinetics in animal and human models are described in detail by Davis et al.
] and Greisler et al.
]. They revealed that the graft healing process represents an intricate, interconnected multicellular process which is the result of a variety of host body response mechanisms; however certain factors of the healing process are highly conservative and determines the total outcome for the restoration of vessel functionality. The graft healing process involves the coordination of host immune cells activity, migration, infiltration, proliferation and differentiation of endothelial cells (ECs), smooth muscle cells (SMCs) and their progenitors which culminate in the formation of new tissue. Their collaborative, balanced regulation and maintenance are pivotal for successful graft implantation. Disruption to the inter-regulation of this process or the negative effect of exogenous factors (bacterial infection, nutrition deficiencies, stresses and a variety of neurological factors) are frequent causes for failures of synthetic grafts by provoking pathological issues such as thrombosis, anastomotic hyperplasia and limited re-endothelisation. [2
Many attempts have been made to increase the patency rates of vessel grafts. Improvements in manufacture, pre-treatment, modification of surface charge and topography, incorporation of therapeutic agents and bioactive coatings, in vitro
addition of a cell monolayer [4
], have all been shown to improve the early stage human healing processes, but have not demonstrated ideal outcomes in terms of long-term performance [4
]. The general concept of material bioinertness (nontoxic and non-antigenic) [15
] for vascular synthetic grafts has been unsuccessful in stimulating positive healing responses and improving the long-term patency of the vascular graft.
An important alternative to synthetic materials is the use of extracellular matrix biomaterials derived from ex vivo
]. To date, numerous animal trials have demonstrated the performance of extracellular matrix (ECM) material for vascular graft applications, highlighting excellent healing properties during implantation for a variety of blood vessels (aorta, artery, vein) [19
]. At certain sites of implantation, the healed graft site exhibited a histological appearance similar to that of normal blood vessel with evidence of early capillary penetration and full endothelisation. There was no evidence of infection, intimal hyperplasia, aneurysmal dilation and the site contained a smooth muscle media and a dense fibrous connective tissue adventitia [22
A thorough elucidation of ECM material characteristics and their influences on the healing process is critical to the understanding of graft pathology and may lead to greater uses of ECM for augmentation and the enhancement of vascular substitutes which are currently available. The main aim of this review is to describe and summarise factors known to date which have been shown to play a beneficial role in modulating cellular events in relation to the healing process. Disadvantages of ECM materials are described in this paper and methods which aim to improve the tissue-specificity, biomechanical properties and total performance of ECM grafts for a variety of cardiovascular applications are presented. This review is aimed at scientists of interdisciplinary fields and describes many unresolved fundamental aspects of vascular graft healing, thus highlighting areas of interest of future research and development.
2. Origin of ECM Material: Source, Preparation, Biochemical Properties, Storage and Commercial Availability
The common goal of all proposed and developed decellularisation methodologies is to efficiently remove all cellular and nuclear material while minimising any adverse effects on the composition, biological activity and mechanical integrity of the remaining ECM. Preservation of the structural components of the ECM allows the generated matrix to provide structural integrity and biomechanical strength for newly developed tissues and enable efficient reseeding [17
]. A thorough review of the various decellularisation techniques has been previously compiled by Gilbert et al.
]. Biochemical techniques of decellularisation include osmotic shock, solvent extraction, ionic and non-ionic detergents, acid/alkaline treatments and enzymatic digestion with DNase, RNase, lipase and proteases. Recently developed methods based on the application of periodic pressure for decellularisation of the skin have been reported in literature [26
]. In addition to these methods, bacterial enzymes produced by Micrococcus luteus
have been investigated for removal of cellular components and their potential in achieving a high quality decellularised matrices [27
]. Protocols vary greatly, and are all known to have limitations [18
Numerous organs and tissues from warm-blooded animals (bovine, sheep, monkeys, pigs and rabbits), including humans, have been investigated to date as potential sources of ECM [18
]. Tissues like the aorta [18
], vein [30
], bladder [18
], small intestinal submucosa [18
], skin [18
], ureter [31
], liver [18
], amniotic membrane [32
] and tendon [33
] have been successfully decellularised and investigated.
ECM materials can be manufactured and obtained in multiple forms. Multi-laminate sheets can be prepared by vacuum pressing methods [18
] and crosslinked by various chemical and enzymatic agents [34
]. Particulate suspensions, powdered and gel forms of ECM can easily be prepared and delivered by minimally invasive techniques to the site of interest [38
], where gel can self-assemble into 3D matrices. Material can be stored in the hydrated state without lyophilisation and cryopreserved [39
]. Lyophilised forms of ECM have a long shelf life and are easily transportable. However, non-lyophilised, hydrated ECMs have shown excellent biomechanical and biochemical characteristics with improved cellular ingrowth rates, illustrating the importance of hydration on properties such as biological effects, degradability, mechanical properties of the structural components (strength and loading response) and volumetric change [40
Naturally derived materials, such as porcine small intestinal submucosa (SIS), urinary bladder matrix (UBM), collagen and fibrogen, offer many mechanical, chemical and biological advantages over synthetic materials, due to their origin and nature [17
]. Their natural three-dimensional ultrastructure and diverse composition of structural and functional proteins including collagen, elastin, growth factors and proteoglycans contibute to overall advantageous properties as a graft material [17
]. However, it must be acknowledged that ECM scaffolds consist of structural and functional molecules secreted by the resident cells of each tissue and organ from which they are prepared. Therefore, the specific composition and distribution of the ECM constituents will vary depending on the tissue source [17
]. The biochemical composition of several materials were evaluated and presented by Badylak et al.
]. In brief, SIS scaffold material has been the most extensively characterized of all ECM materials. It is made up of approximately 90% collagen, the majority of which being type I, with minor amounts of types III, IV, V and VI also present. SIS ECM also contains glycosaminoglycans (GAGs), including heparin, heparin sulfate, chondroitin sulfate and hyaluronic acid. The amount of GAGs present in the tissue after processing depends greatly on the decellurisation methods used [18
]. Previous studies have shown presence of the molecules fibronectin and laminin [44
] which are known to provide adhesion ligands for cells, as well as growth factors [46
] in SIS materials. UBM also contains the same collagen types as SIS but with greater amounts of type III as well as the presence of type VII, which is a central component of the endothelial basement membrane, a distinguishing characteristic of this matrix. Limited research was carried out on characterization and comparison of the spacial distribution of structural domain and the molecular components within the diverse range of ECM matrices. Localization and 3-D positioning of structural and biochemical components within the ECM matrices may have effect on cell adhesion and their migration pattern within the graft. With the use of 3-D imaging fluorescent microscopy, optical sectioning techniques and monoclonal antibody development, further investigation into the parameters mentioned above is possible. This information will help indentify the additional aspects and significance of material structure in relation to the rate of tissue healing.
A partial list of ECM-based, commercially available products is given in Tables 1
. Form, source and composition of intact ECMs commercially exploited to date are diverse. This makes them highly desirable for many clinical applications. SIS is the most extensively clinically used ECM and has been used to reconstruct various tissues such as the abdominal wall, urinary bladder wall, tendons, intestine wall tissue, urethra and ureter [47
]. There are a variety of commercially available collagenous matrices, typically consisting of renatured-purified collagen lyophilised to form sheets or wafers. Examples include collagen type I sheets, collagen matrix (collagen type I and types I and III), connective tissue composites and collagenous gels. Two of the most established products in clinical use are Dermagraft® (Anginera™) and Transcyte®. These are most commonly used in the treatment of diabetic foot ulcers and surgically excised full-thickness and deep partial-thickness thermal burn wounds, respectively. Dermagraft® is a cryopreserved living allogenic dermal equivalent, made from neonatal foreskin fibroblasts cultured on a bioresorbable polyglycolic (Dexon®) or polyglactin (Vicryl®) polymer matrix, and characterised by production of growth factors such as vascular endothelial (VEGF). Transcyte® is a temporary skin cover made from a collagen-coated non-bioresorbable nylon mesh seeded with allogenic neonatal human dermal fibroblasts. Both products are now being investigated for off-label applications with promising results.
5. Vascular tissue Functionality and Homeostasis Maintenance
5.1. Biomechanical Properties of Graft Materials and Their Importance in Sufficient Reconstruction of Vascular Tree
Mechanical factors are recognized as important parameters which affect short and long-term graft patencies [4
]. Researchers have identified that one of the factors limiting the success of prosthetic vascular grafts is material mismatch with the host vessel in terms of strength, stiffness, compliance and elasticity. Salacinski et al
., have shown that mechanical properties including compliance mismatch, diameter mismatch and Young’s modulus all greatly affect the graft performance and failure rate [173
] and support the idea that an ideal vascular graft must have similar viscoelasticity to the native vessel [174
]. Data for compliance and elasticity of native vessels and various small-diameter grafts made from different materials is tabulated in Table 3
The compliance of 3-layered small diameter SIS grafts was shown to be 4.6% (d = 5 mm) and 8.7% (d = 8 mm) [177
], only slightly less than those of carotid and femoral arteries, about four times more compliant than a typical vein graft and an order of magnitude more compliant than modern synthetic vascular grafts of expanded ePTFE and Dacron®. The influence of these particular parameters on flow pattern and hence the cellular healing process will be highlighted in Section 5.2.
Mechanical properties of ECM material vary depending on the animal species, age, organ, physiological and mechanical functions of the organ, size of the organ, species-specific localisation of structural components, protein-homology and protein identity within the tissue and degree of ECM cross-linking [177
]. For example, the modulus of elasticity for canine jejunum is 8.5 MPa [183
] compared to that from porcine sources at 7 MPa [177
]. Manufacturing and processing parameters such as dehydration time, technique, sterilisation methods and storage conditions are important parameters which could potentially affect the mechanical properties of ECM devices and should be taken into consideration when engineering tissue constructs [182
]. The effect of varying the number of layers on the biaxial strength is illustrated in Table 4
For example, sterilisation by ETO has been shown to have the least detrimental effect upon the mechanical properties of UBM. Gamma and e-beam irradiation decrease the uniaxial and biaxial strength and maximum tangential stiffness. However, ETO had no effect on strength or energy dissipated, indicative of unchanged viscoelasticity. All methods significantly decreased material stiffness, when compared to non-sterilised controls (48–60%) [184
In addition, the mechanical properties of ECM material are more dynamic than those of synthetic materials due to the susceptibility of ECM material to degradation and remodeling in vitro
and in vivo. In vitro
studies have shown that material undergoes ultrastructural changes during incubation with Human Umbilical Vein Endothelial Cells (HUVECs) seeded onto the substrates. UBM weight losses up to 2.5% over a 5 day period were recorded. ECM weight loss correlated to an increased production of metalloproteases MMP-1 and MMP-9, both of which are known contributors to ECM degradation, angiogenesis and vessel remodeling [185
]. In vivo
studies provided complementary results and showed that during graft healing and remodeling, the compliance, modulus of elasticity and burst pressure of ECM graft approached the corresponded mechanical properties of native vessel [186
]. Sterilisation of ECM material has been shown to influence on the rate of ECM degradation. An extensive in vitro
degradation study of SIS over a 49 day period found that e-beam irradiation almost doubled the rate of hydrolytic degradation compared with unsterile SIS, gamma irradiation and ETO (42% versus 23–27%) [187
]. These changes are a result of collagen backbone degradation and the difference between the radiation methods could be attributed to the dose and form. Hence, the choice of sterilisation technique should be carefully considered and tailored to the intended application, load bearing requirements and degree of degradability.
In order to engineer a graft which mimics the native soft tissue, manipulation of certain variables such as the origin of the ECM, the number of layers [42
], the decellularisation method (physical, enzymatic or chemical treatment) [18
] and sterilisation techniques (ETO, gamma irradiation, electron beam irradiation) must be determined and elucidiated in order to form a stable reproducable protocol for graft development in each size and substitute location (vein, artery).
The principal strategy being developed to prevent hemodynamic disturbance within the region of the anastomosis is based on the design and fabrication of more compliant ECM–based grafts with viscoelastic properties which mimic those of the human artery. Methods have been introduced to characterize flow structure and wall shear stress (WSS), which may be used in order to provide quantitative comparison of different haemodynamic environments associated with various vascular geometries. Computational fluid dynamics (CFD) and finite element analysis (FEA) softwares may be applied in studies to visualise the flow pattern using velocity vectors, velocity contours and shear stress distribution within the vascular tree or graft replacement region. As these parameters are very difficult to measure in vivo
, computational modeling can become a nessesary and essential tool of analysis the effect of geometry and shape of the graft placement site on flow pattern and severity of flow alteration [188
]. Indeed, CFD and FEA offer much more repeatability and resolution than in vitro
and in vivo
methods, however, computations must be carefully validated against experimental and clinical data. Preliminary evaluations of graft design have utilised CFD to characterise wall shear stress on tubular ECM grafts and FEA has been used to evaluate stress distributions during mechanical testing of ECM materials; both of these computational methods show good prospects for utilisation in the evaluation of ECM materials as a graft material [190
]. The major development of clinical imaging, such as magnetic resonance imaging (MRI) or computed tomography (CT), opens new avenues for detailed patient-specific information on the actual hemodynamics and structural behavior of living tissues. The coupling of CFD/FEA with clinical biomedical imaging technologies may provide an efficient standard evaluation of material performance as a part of clinical practice in the surgical planning and design of graft materials for specific location in the vascular tree.
5.2. Mechanotransduction Pathways in the Healing Process of Vascular Graft
It is desirable that the compliance of the graft matches that of the native vessel to avoid potential stagnant regions [181
] or disturbances to the local haemodynamics around the anastomosed site. This interruption establishes abnormal pulsatile mechanical stresses at the anastomosis. These stresses can result in suture-line disruption, formation of a false aneurysm, and development of subintimal hyperplasia [3
]. These events can either result in thrombosis and increased pannus ingrowth at the anastomosis, thus threatening the patency of the implanted vascular graft. The vascular endothelium is a vital organ, whose healthy physiology and function are essential for normal vascular vessel physiology. The dysfunction of vascular endothelium can be a critical factor in the pathogenesis of vascular disease. ECs lining the blood vessels are transducers of various physiological stimuli which are actively involved in many physiological processes such as regulation of selective permeability, blood coagulation and homing of immune cells to specific sites of the body. Activation of mechanotransductory intracellular pathways is pivotal to shear stress adaptation and is regulated via integrin–mediated connections and cells within the sub-endothelial substrate. These pathways, which are influenced by shear stress, are known to modulate gene expression, cell migration and proliferation. Relationships between hemodynamic parameters [191
], such as wall shear stress and intra-luminal healing, show that a molecular and cellular cascade is triggered by the various flow patterns created at the anastomosis site which reflects the compliance differences between the material and native vessel. The latter results in turbulent flow, which due to its action on vascular endothelium induces pro-inflammatory and pro-thrombotic expression pathways. Molecular determinants of these cellular pathways stimulated within anastomosis sites of synthetic (PFTE) graft are summarised in Table 5
As can be seen from the data, molecular determinants of osteogenesisis and vascular bed remodelling are present in the vascular tissue in the area of turbulent flow, where the molecular marker of SMC contractility (like smoothelin) is down regulated demonstrating that blood vessel function and structure are pathologically altered.
It was shown in a previous study by Orr et al
] that flow-induced activation of the atherogenic transcription factor NF-κβ occurs in a matrix-specific manner. In a study conducted by Jalali et al
], it was found that EC mechanotransduction in response to shear stresses requires the activation of integrins by their specific ligands which are supported, controlled and influenced by formation of new integrin-ligand connections. Through integrin mechanotransduction, shear stress produced by blood flow upregulates genes involved in regulation of apoptosis, cell cycle arrest, morphological remodelling and nitrogen oxide production; which contribute to atheroprotective effects [204
]. Integrins are glycoproteins within the membrane, composed of α and β subunits. To date, 18α and 8β subunits have been identified in mammalian cells [204
], with each of these subunits spanning the extracellular and cytoplasmic domain. The major ECM proteins that interact with vascular ECs include collagen, laminin, fibronectin, vitronectin, and fibrinogen [205
]. Various ECM proteins have been shown to bind to different integrins, and in turn activate different signalling molecules. For example, collagen type I matrix binds to α2β1 and α1β1 integrins in the endothelium [206
]. Laminin tends to bind with α6β1 integrin in ECs [209
]. Fibronectin mainly binds to α5β1 and αvβ3 integrins. Vitronectin and fibrinogen also bind to αvβ3 integrin [205
]. The density and distribution of ECM proteins are known to be controlling factors in the level of integrin-ECM adhesive interaction and play an important role in regulating cell migration [205
]. The cytoplasmic domains of both the α and β subunits interact with signalling molecules and cytoskeletal proteins to regulate cellular events (such as signal transduction, cytoskeletal organisation) as well as regulating cell motility via the modulation of integrin affinity and/or avidity.
The integrin-ligand connection acts as a source of communication, transmitting signals from the ECs to the ECM and vice versa
, coordinating cellular activity. As previously mentioned, the predominant structural component of ECM matrices is collagen; EC attachment to this type of matrix occurs via α2β1 integrins. This interaction inhibits activation and nuclear translocation of NF-κβ and the pro-infammatory molecular cascade under pathological flow conditions. This effect may be beneficial at the early stage of cell repopulation and migration through an anastomosis site, allowing successful tissue reconstruction of the graft without thrombosis developing. Integrin mediated mechanotransduction involves multiple kinases (FAK, c-Src, and Fyn), adaptor molecules (CAS and Shc), guanine nucleotide exchange factors (C3G and SOS) as well as small GTPases (Rap1 and Ras) responsible for activating mitogen-activated protein kinases (MAPKs) such as ERK. In static conditions, the integrins are inactive and signalling does not occur. Shear stress is required to activate the integrins. Through specific interactions of the α and β subunits, the FAK/c-Src and Cav-1/Fyn pathways are activated. Activation pattern is directly associated with members of the Rho small GTPase family, including RhoA, Cdc42, and Rac. RhoA, Cdc42 and Rac each have particular functions in regulating the actin-based cytoskeletal structure. RhoA increases cell contractility, focal adhesions and actin stress fiber formation; Cdc42 regulates filopodia formation; and Rac regulates membrane ruffling [210
]. The importance of the β2-integrin family in lipopolysaccharide (LPS) stimulation was highlighted by Monick et al
]. LPS stimulation plays an important role in regulating the inflammatory process, thus the link between this stimulation and the role of β2
-integrins is important in terms of tissue remodelling. It has also been found that laminar shear stress suppresses the G1-to S-phase transition in ECs [212
]. This leads to an increased expression of p21 which inhibits cyclin-dependent kinases, thus inhibiting cell proliferation and remodelling. The regulation of transcription factor expression was compared under disturbed flow conditions and uniform laminar shear stress conditions by Nagel et al
]. The ECs subjected to disturbed flow, similar to that found in atherosclerosis-prone areas, showed increased levels of nuclear localized NF-κβ, Egr-1, c-Jun and c-Fos compared with those exposed to uniform laminar shear stress or under static conditions. NF-κB induces the transcription of a large range of genes implicated in inflammatory response [215
]. It also plays a fundamental role in protecting vascular SMCs against apoptosis and weakening of vascular wall. This transcription factor is stimulated by flow through the integrin and Rac dependent production of reactive oxygen species. It has been previously shown that uniform laminar shear stress plays an inhibitory role in the pro-inflammatory gene expression in ECs located in close proximity to SMC [214
As the specific activation of inflammatory cascades is complex, further studies into the gene expression of various scaffold materials are needed to predict in vivo performance. Intuitively it would be expected that the performance of synthetic grafts would be impeded due to their nature and the inability of integrins to bind to corresponding ECM ligands and thus inhibiting of initiation of “healthy” mechanotransduction within the vascular cell. From this point of view, naturally derived scaffolds have a distinct advantage. However, certain ECM components trigger more favourable responses than others. Gene expression in response to physiological fluid flow has yet to be fully characterised for all ECM materials.
In a study conducted by Cenni et al
], integrin expression was evaluated for ECs alone and ECs in contact with polyethylene terephthalate (PET) woven Dacron. The following integrins were evaluated under both conditions by flow cytometry: VLA-2 (α2β1-CD49b/CD29), VLA-5 (α5β1-CD49e/CD29), VLA-6 (α6β1-CD49f/CD29) and αVβ3-CD51/CD61). The isolated ECs in contact with woven Dacron showed a significant decrease in the expression of CD29 and CD49e and the other integrins were not modified by contact with the material. CD29 and CD49e are the α5β1 integrin types which are known to have affinity to fibronectin ligands and are important for cell adhesion. The decrease in this integrin type may suggest that adhesion chemistry between vascular cell and synthetic material differs from that of cell adhesion to natural ECM components. Strength of adhesion and retention under the flow shear stress and rate of EC migration on this type of substrate may lead to incomplete endothelial lining which may result in thrombogenicity [216
]. To combat these issues, synthetic materials such as PET and PTFE are often coated in fibronectin. Plasma treated PET and PTFE have also shown improved adhesion and growth of ECs [217
]. Depending on the nature and origin of ECM materials, different integrins are expressed. Activation of certain integrins suppresses the activation of other specific integrins, maintaining a balance which aims to promote healthy tissue remodelling. Manipulation of integrin activation could eliminate adverse remodelling events, for example, by blocking the inhibition of integrin α2β1 (activated on fibronectin through protein kinase Cα) then collagen signaling can occur via this integrin and inhibit the flow induced activation of the atherogenic transcription factor NF-κβ [218
]. The state of EC surface thrombogenicity is under substrate control, and is also related to the cellular differentiation status (as shown in Figure 3
These cellular processes demonstrate the potential of the underlying vascular material to affect the long-term cellular functionality of the prosthesis. Ex vivo
evaluation of the material properties in order to support functionality of the EC and their mechanotransduction capacities require optimization. In vitro
studies under static conditions have been popular for characterization of EC thrombogenicity and shear resistance due to their logistical simplicity, but are not necessarily reflective of the surface thrombogenicity under flow conditions where mechanotransduction activation in a fashion similar to the vascular tree in vivo
are necessary and its modulation by substrates may be analysed and determined. A biorheological conditioning protocol proposed by O’Keeffe et al.
] can provide a efficient informative in vitro
screening method which would allow determination of vascular cell behaviour on surface graft material under pathological shear stress, elucidate an alteration pattern of the molecular cascade of endothelial cell mechanotransduction, investigate the ability of materials to support a polarization and EC cytoskeleton reorganisation under various flow patterns (which may be recreated in the anastomosis site of a graft vessel). The undestanding of molecular mechanisms of graft failure will lead to the possibility of further modification of the vascular material in order to enhance clinical performance of prosthetic and ECM grafts.
5.3. Restoration of Innervation and Blood Vessel Homeostasis
In recent years, biomedical research has clarified the involvement of neuromodulation in human tissue processes which occur during healing [220
]. However, limited data exists on the re-innervation pattern of vascular graft materials during the healing process due to the complexity of the neuron detection and analysis [222
]. However even limited findings have demonstrated the role of innervation and its pattern may play a part in the development of fully functional tissue during the healing of vascular graft [224
Anatomical investigations reveal that blood vessels and nerve fibres run throughout the body alongside one another and the mechanisms involved in wiring both networks are proposed to be similar (Figure 4
]. Recent morphological and pharmacological findings support the hypothesis of active communication between vascular and neural networks and their interactions may contribute to the health and homeostasis of vascular vessels [229
The blood vessels are innervated by the autonomic nervous system. Sympathetic adrenergic nerves, which travel along arteries and nerves, are found in the adventitia (outer wall of a blood vessel). The sympathetic fibres mediate a vasoconstrictive action in the vascular bed as well as providing secretomotor activity. Activation of vascular sympathetic nerves cause vasoconstriction of arteries and veins mediated by α-adrenoreceptors [232
]. Neurogenic control of vascular tone and vascular innervations of the blood vessel adventitia have been well documented [230
]. The release of neurotransmitter and chemical signalling occurs in small enlargements along the nerve fibres. Nerve stimulation can elicit different responses, in terms of type and amplitude, at different areas of the vascular system [234
]. Neurogenic vasocontrol can thus follow different patterns depending on which molecules are released and what local reactions they trigger. Active molecules released locally from adventitial nerves may diffuse and act directly on the adventitia and the media or act on the endothelium which in turn will release molecular signals (like nitrogen oxide) which then influences the media [229
]. The complexity of neurochemically defined autonomic nerves stimulating the vessel baroreceptor and chemoreceptor regions suggests functionally separate, independently regulated pathways. Auger et al.
] summarised previous literature and concluded that vascular autonomic and various types of sensory nerves inclusive of cholinergic, adrenergic, peptidergic or nitrergic are found in different proportions and density depending on the specific anatomical site. One of the findings of this study was that the variety of nerves which can be found in the adventitia is directly related to the presence of many neuron-related peptides and molecules (such as acetylcholine, noradrenaline, neuropeptide Y, substance P(SP), calcitonin gene-related peptide (CGRP), neurotensin and vasoactive intestinal peptides (VIP)) in different quantities at various anatomical regions, although their complete physiological roles are not yet known.
Modern surgical procedures utilized for implantation of vascular grafts have been shown to cause extensive damage to the sympathetic nerves which supply and accompany blood vessels [237
]. Some procedures may cause extensive degeneration of adrenergic nerves and the extent of denervation may vary with vessel type with regard to their anatomical characteristics and structure (elastic or muscular). Comparative studies were able to determine that rate of nerve re-growth in muscular vessels is faster than that of elastic vessel. Re-growth of injured fibers can be altered and lead to hyper or denervation along the graft reconstruction or only at certain parts. Preliminary analysis of data indicates that a non-matching to the original pattern of nerve re-growth may lead to a lack or alteration of vessel tone autoregulation in the graft region and its exclusion from the baroreflex modulation of blood flow. This growth pattern is potentially regulated by a number of factors such as chemical affinity of the material surface, cytokine and growth factor gradient along the grafted site, the type of ECM molecule deposition, adsorption to the material, the presence and distribution of required chemical ligand to the surface and within the scaffold, inflammatory sites and infection.
In newly developed vascular tissue, neuromodulation activity or its complete absence appears to greatly affect the functionality of the vessel itself, the healing process of the vessel and its homeostasis. Instability, due to focal unbalancing of constrictive forces and regulated, molecular signaling may serve as a pathophysiological basis for the well-established phenomenon of vascular SMC hypertrophy and hyperplasia after grafting [239
], occlusion of the vascular graft [240
] or its dilation (pseudo aneurysms) [241
Recent studies have reported successful re-innervation of ECM reconstructed organs [242
], and the capacity of ECM materials to support nerve conduits and promote growth of the Schwann cell (SC) [244
]. During biocompatibility studies in vitro
], when co-cultured with SCs, SIS-ECM showed good ability to support SCs adhesion, survival, migration and proliferation on its surface. Observation of the ultrastructure of SCs by TEM demonstrated that SCs adhered tightly and grew productively on the surface of SIS. MTT assay also showed that SIS did not have a cytotoxic effect on SCs. Quantitative analysis of nerve growth factor-β (NGF-β) and brain-derived neurotrophic factor (BDNF) by ELISA, as well as semi-quantitative analysis of NGF-β mRNA and BDNF mRNA by RT–PCR, showed that SCs seeded on SIS had more productive function of secretion than the normal cultured SCs. NGF-β and BDNF are the main growth factors secreted by SCs, which are known to have neurotrophic effects on nerve regeneration. Adhesion and growth of the nerve cells is guided by specific structural components of ECM laminins [249
]. There is a significant lack of research focusing on the analysis of innervation fibre density, their localization in the adventia and media of the synthetic and ECM grafts. This issue needs to be elucidated and determined in future studies, but the authors hypothesize that ECM graft material may provide a better support for the re-innervation of the newly developed vascular tissue and an enhanced structural and chemical microenvironment for the reconstruction of a balanced interactive network between the nervous system and remodeled vascular tissue. ECM material has the potential to provide a homeostatic environment based on the regulation of vasoactivity function, thus increasing its ability to regulate blood flow and function in a similar manner to the native non-injured vessel compared with vessels substituted by synthetic materials.
6. Future Perspectives for Cardiovascular Implants Based on ECM
With the enhanced healing properties previously discussed, biologically derived ECM materials have been shown to be diverse and inconsistent in both structure and morphology [41
], and affected by a number of factors such as the manufacturing process (i.e.
, mechanical decellularisation vs. chemical decellularisation) [18
] and the age and health status of the animal at harvest. Numerous limitations and problems are associated with decellularising techniques and procedures, as described previously in detail [18
], may have great influence on ECM product quality, mechanical and biochemical properties, biocompatibility and clinical performance [257
]. Mechanical methods of acellularization, including repeated freeze-thawing, sonication, or other physical means of disrupting cells’ plasma membranes, provide a direct, mild and rapid tissue decellularisation, but used alone, such methods are not capable of completely removing cellular material, which has been shown to prevent complete recellularisation of this material by host cells [257
]. It remains a concern for the biomedical community that trace amounts of potentially antigenic compounds of animal origins (lipids, DNA, glycosilation products) have been reported to be present for certain types of ECM material and may provoke an inflammatory response at the placement site [260
]. Complete removal of these antigenic compounds from the material is important for improving the biocompatibility of ECM material; however, such a goal seems to be quite challenging and difficult to achieve with the application of standard biochemical extraction methods, due to the high degree of chemical complexity and variability of the contaminants. The degradation and damage of structural components due to the absence of a long-active protease inhibitors, multiple incubation and rinsing steps during the long decellularisation procedures may unintentionally remove desirable ECM components and lead to an alteration of mechanical properties of the ECM material [262
]. The ability of a decellularisation method to sufficiently remove of lipid moeities from the tissue has been shown to have a dramatic impact on the rate of graft calcification and substantially decrease patency time in vitro
]. Minimally invasive tissue-specific decellularisation techniques with detailed manufacturing protocols where in vivo
performance will be supported by enhanced cell repopulation, minimal infiltration of inflammatory cell and calcification still need to be developed in the near future.
Suitability of materials derived from various organs for clinical application in the vascular area are still under investigation as the biochemical and biophysical properties are not always tailored to the potential tissue characteristics desirable for the application [17
]. Modification, remodelling and ECM matrix deposition that can be performed in vitro
utilising cellular machinery able to perform the most complex synthesis reactions and cleavage of others with high specificity and efficiency, presents a potential method to improve the biological properties of ECM material obtained by decellularisation of non-vascular organs (SIS) for vascular application [265
]. SIS-ECM material was pre-conditioned with human endothelial cells in vitro
and during the conditioning process, remodelling of the matrix and synthesis of novel components as well as deposition of sub-endothelial matrix (basement membrane) known to be absent in the original SIS material, was shown to occur. Following decellularisation of the cell-seeded scaffold, neo-ECM was shown to have improved biological activity and the vascular endothelial cells seeded on the neo-matrix had enhanced organization of the cell junction, an increased metabolic activity and released a lower amount of pro-inflammatory prostaglandin PG1 compared to the cells incubated on the control SIS. Neo-ECMs were also shown to have a lower degree of human platelet adhesion and improved thrombogenic potential.
In this state-of-the-art era, with strong development in genetic engineering methods [266
] and the multitude of recombinant vectors [267
], gene delivery systems [268
], the variety of cell lines which are able to express recombinant ECM molecules, enzymes and growth factors [281
] all show great promise for ECM material modification. A more complete understanding has emerged of the native molecular regulation of the ECM remodelling process by the cells under chemical and/or mechanical stimulation (flow [284
], stretch [285
], pressure [286
]), providing an optimised physiological environment within an engineered bioreactor. This may facilitate a more physiological remodelling process, ultimately leading to a more manufacturable tissue-specific ECM material. Advanced bioreactor design and application will be an essential part of conditioning the ECM material and ECM-material based constructs with further advances in the technology making the engineering of more complex tissues and organs (multi-functional, multi-layered, bio-chemical, bio-properties) a reality in the clinical environment.
ECM material has been shown to a very attractive material as a base or one of the structural components for a composite repair material as part of the continuing challenge to find ways to translate the mechanical properties and clinical performance of ECM biomaterials to vascular clinical applications. In order to increase the mechanical strength of ECM, cross-linking structural components with chemicals such as glutaraldehyde, 1-Ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC or EDAC) and hexamethylene-diisocyanate is applied; however, modification by this type of method has been shown to have a lower degradation rate in vivo, promote early calcifications and changes the host tissue response from an anti-inflammatory, constructive remodelling response to a pro-inflammatory, foreign body response.
Decellularised ECM materials have been incorporated with synthetic scaffolds successfully to date through a number of approaches. A multilayered poly(styrene sulfonate)/poly(allylamine hydrochloride) (PSS/PAH) have recently been used as luminal coatings onto human umbilical arteries, demonstrating a high graft patency post 3 months rabbit implantation and restoring initial compliance of the tissue [287
]. Stankus et al
. developed a composite scaffold containing poly (ester urethane) urea elastomer (PEUU) and UBM [288
]. As the water-soluble electrospun UBM was a fragile and brittle material, it was blended with PEUU and dissolved in hexafluoroisopropanol and then electrospun. The scaffold was also more resistant to degradation compared to electrospun UBM alone and had improved mechanical properties. The scaffold was both strong and distensible with a tensile strength of 4.9 ± 1.6 MPa and a breaking strain of 85 ± 28%, compared to lyophilized sheets of UBM (0.3–0.4 MPa and 47–67% strain, respectively) [288
]. After 28 days implantation in a rat subcutaneous model, there was an increase in scaffold degradation and cellular infiltration with increasing UBM proportions. In vitro
seeding with SMCs displayed enhanced adherence and proliferation with increasing UBM proportions. Most likely, this is due to increased cell adhesion sites retained from the biological component (e.g.
, collagen, fibronectin, laminin) and growth factors, which survived the enzymatic digestion and acidic conditions during initial processing and electrospinning solvent conditions. Despite the fact that the harsh fluoroalcohols typically employed to electrospin collagen/ECM material effectively denature collagen to gelatin [291
], with losses of more than 90% of triple-helical structure, the gelatinous structures with retained growth factors do confer an enhanced bioactivity. The synthetic portion comprising of bioresorbable or non-resorbable meshes such as Prolene™, Vicryl™, Mersilene™, PDS II™, Panacryl™, and Monocryl™ can be introduced by preparation of laminated structures of ECM sheets presenting a new method of manufacturing hybrid-ECM material with enhanced properties. Hence, a synthetic scaffold enhanced with an ECM component may possess more consistent mechanical properties, such as failure strengths, compliance and degree of shrinkage [292
], whilst eliminating the need to cross-link or laminate the structures. Incorporation of ECM components into the synthetic material adds several desirable characteristics which are absent even from the most advanced textured synthetic scaffolds and forms a ‘smarter’ biomaterial [293
], and introduces new functions of the material with near-physiological multifunctionality of the natural ECM, complex signalling, improved control of cell-matrix interactions [294
] and cell-specific matrix response.
Specific ECM components have been commonly used (e.g.
, collagens, fibronectins, laminins) in cell culture for many years and have been shown to have strong effects on cell attachment and growth. The intricate, ordered nature of the ECM, combined with the complex combination of biomolecular cues, is highly difficult to reproduce with synthetic scaffolds. At best, common synthetic ECMs exploit one or two biomolecular classes. A recent study elegantly demonstrated that the tissue-specific matrix components cause significant differences in adhesion efficiencies, growth rates, morphology and phenotypes of skin, muscle and liver cells, suggesting the need for more appropriate, tissue-specific matrices for in vitro
cell culture [296
]. As mentioned previously, the major disadvantage of synthetic vascular materials is their lack of a confluent endothelium and that they are prone to thrombus induction, embolism and occlusion. They are also less durable than autologous material and are associated with poor healing and lack of compliance and often require extensive use of anticoagulant or antithrombotic agents [297
]. Therefore, the importance of creating a suitable endothelium on the luminal surface of any diameter synthetic vessel substitute is paramount. Coatings of proteins, decellularised matrices are being pursued to increase the bioactivity of the prostheses in order to render then more suitable for EC seeding. Common approaches to treat vascular graft surfaces include autologous fibrin coating [298
] or heparin [300
] and have been met with mixed success. On the whole, these linings do not provide vital vascular functions such as vascular responsiveness or other biological secretory functions seen with normal blood vessels [302
]. A mixture of collagen type I, elastin and poly (D, L-lactide-coglycolide) (PLGA) to impart increased mechanical strength was electrospun to form a non-cytoxic tubular construct with similar compliance to native arteries and minimal inflammatory response [303
]. Tillman et al.
electrospun a collagen type I-PCL tubular scaffold, pre-seeded with ECs and SMCs under bioreactor conditioning flow [304
]. The grafts used had a caliber of 5 mm and were able to support EC and SMC growth under pulsatile flow conditions. Although smaller diameters are associated with a higher degree of thrombotic occlusion most remained patent at 1 month, even without the presence of anti-thrombogenic ECs. For example, the Hemashield™ vascular graft is a woven double-velour polyester graft impregnated with weakly cross-linked purified bovine collagen, softened by exposure to glycerol. This confers anti-thrombogenicity, an improved healing response and eliminates the need for pre-clotting, which is patient-specific, time-consuming and troublesome.
Replacement of the vessel, in major surgical cases, may serve as a symptomatic treatment of atherosclerotic lesions or aneurisms that have developed in the vessel wall altering its structure and function but does not target the cause of disease known to include molecular and biochemical processes in the blood and the blood vessel wall, and their interface such as deposition of plaques, cholesterol, platelets and other related molecules within the arterial walls. Due to this factor, long–term performance of the ECM material and functionality of the new vessel tissue depends on the origin of the disease, genetic predisposition, specific diet, age, hormone status and many other risk factors. ECM as a protein–based material may be further modified and activated as a potential delivery method of therapeutic agents (drugs, enzymes, inhibitors) which can be released after incorporation of the material into the body. This approach may only provide a short-term treatment due to quick exhaustion of quantity and activity of the loaded functional protein or drug. Incorporation of the specific DNA molecules with coding sequence of the disease-related gene, its uptake by the host cell during repopulation of the ECM vessel following development of the stable/induced subpopulation of the vascular cell may potentially provide a new resistant to arteriosclerosis development vascular tissue due to continuous or inducible synthesis appropriate enzyme, cellular receptor or ECM components coded by delivered DNA molecules.
The future perspectives for the application of ECM technology to vascular applications still have many obstacles to its success and development, with decellularisation, material properties and component modification processing still at the development stages. The advancement of these technologies should ensure a highly improved bioscaffold for the treatment in cardiovascular applications. Furthermore with the application of new research and approaches to ECM materials may allow many of the outlined shortfalls in the current treatment approaches to be eliminated.