Vascular tissue engineering: biodegradable scaffold platforms to promote angiogenesis

The ability to understand and regulate human vasculature development and differentiation has the potential to benefit patients suffering from a variety of ailments, including cardiovascular disease, peripheral vascular disease, ischemia, and burn wounds. Current clinical treatments for vascular-related diseases commonly use the grafting from patients of autologous vessels, which are limited and often damaged due to disease. Considerable progress is being made through a tissue engineering strategy in the vascular field. Tissue engineering takes a multidisciplinary approach seeking to repair, improve, or replace biological tissue function in a controlled and predictable manner. To address the clinical need to perfuse and repair damaged, ischemic tissue, one approach of vascular engineering aims to understand and promote the growth and differentiation of vascular networks. Vascular tissue engineered constructs enable the close study of vascular network assembly and vessel interactions with the surrounding microenvironment. Scaffold platforms provide a method to control network development through the biophysical regulation of different scaffold properties, such as composition, mechanics, dimensionality, and so forth. Following a short description of vascular physiology and blood vessel biomechanics, the key principles in vascular tissue engineering are discussed. This review focuses on various biodegradable scaffold platforms and demonstrates how they are being used to regulate, promote, and understand angiogenesis and vascular network formation.

bio mecha nics, the key principles and cell sources for vascular tissue engineering are discussed.

Vasculogenesis and angiogenesis
During embryonic growth, new vasculature develops through vasculogenesis. Angioblasts diff erentiate into endothelial cells (ECs), which cluster to form a tube-like structure supported by smooth muscle cells (SMCs) [10]. ECs create the selectively permeable lining of blood vessels, forming a barrier that resists thrombosis and facilitates platelet activation, especially during wound healing. By producing collagen and elastic fi bers, SMCs provide contractile and elastic forces, which support blood vessel integrity. After initial blood vessels form, the vascular network continues to grow through a process called angiogenesis, which is particularly important during natural wound healing and also during cancerous tumor survival. Th e extracellular matrix (ECM) has a diverse composition that helps regulate angiogenesis by providing critical signaling cues, EC receptor interactions, and the reten tion of growth factors [12][13][14][15][16][17]. During this process, proteases degrade the ECM to make way for new vessel formation.
In angiogenesis, vessel branching generally occurs in three stages: quiescence, activation, and resolution [10]. During quiescence, EC proliferation is inhibited as ECs are tightly interwoven with vascular endothelial cadherins and are supported by pericyte cells. Activation usually occurs when a vessel receives angiogenic signaling cues, such as vascular endothelial growth factor (VEGF), from another cell source. Upon activation, pericytes break away from the basement membrane. Th e basement membrane degrades, allowing room for extending ECs to migrate [10]. Th e EC monolayer dilates as the vessel's permeability increases by VEGF signaling, and cell junctions become less tightly bound. A tip cell, an EC with fi lopodia that is chosen to sense the microenvironment, leads the direction of vessel formation. Th is tip cell extends from the degraded basement membrane with the help of directional cues from angiogenic factors [10,16,18]. Th e surrounding ECs are known as stalk cells, which support the tip cell, proliferate to lengthen the extending stalk, and eventually form a new vessel. During resolution, the extending tip and stalk cells fuse with another extending vessel branch. EC junctions are reformed, and pericytes reattach to newly laid basement membrane [10].

Mechanical forces and oxidative balance
Blood fl ow and pressure act on the blood vessel wall to maintain homeostasis through biochemical pathways and mechanical forces. Wall shear stress and circumferential wall stress and strain are the main forces associated with vascular wall biophysical regulation [27,28]. Wall shear stress results from the frictional force of blood fl owing past the EC layer. Circumferential wall stress and strain (stretch) in the circumferential direction result from pressure. Th is pressure is generated by pulsatile blood fl ow and acts perpendicular to the EC layer [28]. In physiological and pathological states, the vasculature can be dilated and remodeled by changes in blood pressure and fl ow.
Oxidative balance is key to maintaining healthy vascular function and homeostasis. Blood pressure causes vessels to stretch beyond their relaxed state, known as mechanical distention. Shear stress caused by blood fl ow activates integrins on the EC monolayer and induces vasodilation. Integrin activation leads to endothelial nitric oxide synthase phosphorylation. Activated endo thelial nitric oxide synthase produces nitric oxide, which stimulates vasodilation, relaxes SMCs, and decreases blood pressure [27,28]. To counterbalance vasodilation and induce vasocon striction, circumferential stretch leads to nicotinamide adenine dinucleotide phosphate oxidase activation that generates superoxide, increasing free radical levels [28]. Free radical anions react with nitric oxide to create peroxynitrite, an oxidant. Th e decreased levels of nitric oxide reduce vasodilation. Oxidative balance between free radical species (oxidants) and antioxidants, such as nitric oxide, controls the vaso dilation and homeostasis of the vascular wall [28]. In tissue engineering, this balance is important to take into consideration when designing solutions to repair vascular damage.

Vascular tissue engineering: cell sources for regenerative medicine
In vascular regenerative medicine, there are two focuses: forming artifi cial blood vessels, and producing tissue constructs that regulate the growth of new vascular networks. Both of these approaches to repair, improve, and understand the human vascular network are founded in the principles of tissue engineering. Generally, the components used in vascular engineering are a biodegradable scaff old, cells from either an autologous or an allogeneic source, and growth factors necessary to create a stimulating microenvironment, as depicted in Figure 1 [7,9,29]. Many grafts and constructs are also preloaded in vitro by mechanical stimulation in a bioreactor, which mimics physiological conditions [1,7,8]. Researchers use various combinations of these compo nents to try to recapitulate human vascular function.
Cell sources for tissue engineering can be divided into three categories: somatic cells, adult progenitor and stem cells, and pluripotent stem cells (PSCs). In these categories, there are numerous cell types that are used for vascular tissue engineering. For further details please refer to current reviews by Bajpai and Andreadis [30] and Reed and colleagues [31]. Briefl y, some common cell sources used for vascular constructs are ECs, SMCs, endothelial progenitor cells (EPCs), mesenchymal stem cells, and PSCs [30,31]. For mature vascular cells, ECs and SMCs can be derived autologously, directly from a patient. Th e use of autologous cells can be ideal for vascular engineering because there is no immunogenic response or cell rejection upon implantation. However, mature vascular cells are terminally diff erentiated with limited proliferation capacity and thus limited expansion ability [8,9].
Adult progenitor cells have more proliferation potential and plasticity to diff erentiate down a specifi c lineage. EPCs can be isolated autologously from peripheral blood and bone marrow [11,32,33]. However, these cells have limited self-renewal capabilities compared with stem cells, and their origin and regeneration capacity are debated. Adult stem cells, such as mesenchymal stem cells, are an autologous multipotent cell source that have high pro liferative capacity, can diff erentiate into SMCs, and have been suggested to be able to diff erentiate into ECs [30,[34][35][36][37][38][39]. Nevertheless, autologous adult pro genitor and stem cell populations can be sparse and diffi cult to detect and isolate. As such, methods for isolating and expanding autologous EPCs and mesenchymal stem cells are generally time intensive and expensive [9]. PSCs, including induced PSCs and embryonic stem cells (ESCs), can diff erentiate into all three germ layers. Th ey have an unlimited ability to self-renew, making them easy to expand for therapeutic use [40,41]. ESCs are derived from a developing embryo, while induced PSCs are generated by the reprogramming of somatic or adult progenitor and stem cells. Allogeneic cell rejection is therefore a consideration when developing ESC-based therapeutics, while induced PSCs hold the potential to be a useful autologous cell source [40]. Human PSCs have been successfully diff erentiated into mature and functional vascular ECs and SMCs [30,31,[42][43][44][45][46][47][48][49][50][51][52][53][54][55][56]. Th erapeu tically, the use of human PSC vascular derivatives has oncogenic concerns, such as teratoma formation due to proliferative or undiff erentiating cell populations [56,57]. Allogeneic cells either from healthy donors or from animals can make cells available via an off -the-shelf route, as cells can be expanded beforehand in large quantities. However, there are problems with graft and construct rejection due to the foreign allogeneic cells, as well as diff erences between donor and recipient cell characteristics such as age, antigens, and proliferation potential.

Scaff old materials
Th e scaff old component is widely used in tissue engi neering, especially to promote and regulate angiogenesis. Scaff olds were originally incorporated to give transplanted cells and the host's regenerating tissue a threedimensional support structure [8,9]. Th e scaff old mimics an in vivo cellular microenvironment better than a twodimensional monolayer, which is a common cell culture method in vitro. Researchers use scaff olds not only as a support for cell growth and diff erentiation, but also as an anchor to attach diff erent bioactive molecules and signaling cues that enhance specifi c cell function. In the case of angiogenesis, molecules such as VEGF can be bound to scaff old surfaces, presenting pro-angiogenic signals to the surrounding tissue [23]. Among the diff erent types of scaff olds, injectable scaff olds are a promising approach for promoting angiogenesis since they are less invasive than surgical implantation and can mold into oddly shaped structures to fi ll cavities and areas of necrotic tissue [58][59][60]. Th is review will focus on pre-formed or pre-constructed scaff olds to promote angiogenesis, but more information on injectable scaff olds can be found in Hou and colleagues [60].
A variety of materials are used for scaff old preparation, including synthetic polymers and derivatives of natural proteins. Synthetic materials are generally reproducible, cheap to fabricate, and readily available. Th is would make synthetic materials a probable therapy to translate clinically. Also, synthetic materials off er researchers control over many critical properties, such as the degradation rate and elasticity. Ideally, synthetic materials can be designed to degrade and resorb into the body at a rate that matches tissue regeneration and growth. However, a common problem with synthetic materials is that their degradation products can be toxic or can cause infl ammatory responses, limiting scaff old success in vivo [9]. Natural-based scaff olds are generally derived from ECM compo nents, such as collagen, fi bronectin, and hyaluronic acid (HA). Researchers use scaff olds made from a single isolated ECM protein, combinations of ECM proteins, and decellularized ECM that was deposited by cells or extracted from a tissue sample or intact organ section [16,17,[61][62][63][64][65][66]. Since ECM components naturally occur in the human body, ECM-based scaff olds support cell attachment, growth, and diff er entiation. Th ey generally do not have harmful degradation products, making it easier to integrate with the body. However, with natural ECM-derived scaff olds, researchers have limited control over material properties such as the degradation rate, strength, and elasticity [9].

Biodegradable polymer scaff olds: synthetic polymers
Biodegradable scaff olds attempt to mimic numerous physical environments in the body. As such, they are designed to present signaling molecules and mechanical cues to cells and surrounding tissue, supporting cell growth, diff erentiation, and proliferation. Synthetic polyesters -such as polylactic acid, polyglycolic acid, poly(lactic-co-glycolic acid) (PLGA), and polycaprolactone (PCL) -are used extensively as scaff old materials [9,21,24,[67][68][69]. Th ese polyesters are usually inexpensive to produce, nontoxic, and degrade by natural hydrolysis in the body. Synthetic polymers can be synthesized with desired properties such as the degradation rate. Th is control makes possible the design of a scaff old that degrades at the same rate at which cell growth and tissue regeneration occur. However, synthetic polymers are limited in their ability to reproduce the complexity of the physiological, cellular microenvironment, as many biological components need to be added to replicate ECMdriven signaling.
Many researchers observe vascular network assembly using a three-dimensional, synthetic polymer scaff old to stimulate seeded cells. Lesman and colleagues co-cultured cardiomyocytes diff erentiated from human ESCs, fi broblasts, and ECs in a porous 50% poly-L-lactic acid (PLLA) and 50% PLGA scaff old mixture to create a beating, prevascular ized, muscle construct for application in myocardial infarctions [2,68]. Th e glycolic acid in PLGA decreased the degradation time of the scaff old, while PLLA provided an appropriate mechanical rigidity for cell culture. Th e polyester scaff old created a unique platform that allowed for successful vascularization and organization of syn chronized, beating, cardiac muscle tissue. Later, Lesman and colleagues combined the 50:50 PLLA and PLGA scaff olds with a fi brin gel, which fi lled the scaff old's pore spaces [61]. When seeded with human umbilical vein ECs and fi broblasts or with human umbilical vein ECs, fi broblasts, and skeletal myoblast cells, this scaff old-gel mixture allowed for interconnected vessel-like network formation in vitro. Th e fi brin gel alone was not as successful because cell forces caused the softer gel to eventually shrink. Th ese studies provided a unique fi brin, PLLA, and PLGA mixture for a scaff old that could successfully support vascular network formation. Des Rieux and colleagues combined nanoparticle technology with Matrigel™ hydrogels or with PLGA scaff olds [19]. An increase in angiogenesis was observed when encapsulated VEGF was incorporated into the PLGA scaff old, increasing local VEGF release. Th is study is one example of many approaches utilizing nanoparticle technology for vascular regeneration. Such approaches are aimed at targeted delivery to the site of injury followed by local release of pro-angiogenic factors, for the effi cient localized retention of the therapeutic agent.
Singh and colleagues established a porous PCL scaff old platform with immobilized heparin on its surface [23]. Heparin's negatively charged sulfate groups attracted and bound VEGF's positively charged amino acids, leading to increased retention and absorption of VEGF in the scaff old. Th e heparin-PCL scaff old had high vessel density and increased endogenous angiogenesis upon implantation in NOD-SCID mice due to better retention and local VEGF delivery. In a following study, Singh and colleagues seeded human EPCs into heparin-PCL scaff olds and observed anastomosis of human EPC-formed vessels with mouse host vasculature after 7 days of subcutaneous implantation [24]. Th is platform improved growth factor retention and decreased leaching, utilizing heparin's negative charge properties. Th is approach thus holds the potential to alter other materials toward angiogenicpromoting properties.

Biodegradable polymer scaff olds: natural polymers
Natural polymer scaff olds are used because of their biologically recognizable side groups, which make them more compatible upon implantation and more likely to support cell function. Th eir composition, compatibility, porous structure, and mechanical properties make them suitable scaff old materials to mimic the natural ECM. Tengood and colleagues created a hollow, porous scaff old from cellulose acetate in the shape of a fi ber that penetrated an in vivo site [21]. Th e scaff old's unique structure and pore size allowed for in vivo basic FGF and platelet-derived growth factor sequential delivery to surrounding tissue, allowing novel study of temporal growth factor release. Th e scaff old demonstrated that sequential delivery was key to EC and pericyte cell colocalization in maturing vessels. Th is platform can be applied to many other biomolecules and used to study the timing on their release and consequences in vivo.
Our laboratory has shown that the natural polymer dextran could be modifi ed with various functional groups and crosslinked with polyethylene glycol diacrylate to form a biocompatible, hydrogel scaff old [70]. Dextran is a nontoxic polysaccharide made of linear α-1,6-glycosidic linkages of d-glucose [70]. Subsequently, dextran's ability to promote angiogenesis was explored. Th e crosslinking density of dextran was decreased, which promoted tissue ingrowth, increased hydrogel swelling, and released more VEGF [71]. Immobilizing a combination of pro-angiogenic growth factors yielded eff ective formation of functional vessels. Th is study showed that such a platform could be a promising clinical therapy. Finally, we applied the dextran-polyethylene glycol diacrylate hydrogel platform to a murine burn wound model, as depicted in Figure 2 [72]. Th e hydrogel scaff old facilitated infi ltration of angiogenic cells, which led to endogenous neovasculari zation and angiogenesis in the wound. Th e results showed an improved wound healing response and accelerated skin regeneration when compared with a bovine collagen and glycosaminoglycan matrix, which is a current treatment for burn wound injury. Th e dextran-poly ethylene glycol diacrylate hydrogel could potentially provide an improved clinical solution to current treatments.

Extracellular matrix-derived scaff olds
ECM-derived scaff olds are optimal for cell attachment, growth, and signaling. Th ey present ECM receptors and promote binding interactions that cells naturally en counter in the body. ECM-derived scaff olds are biocompatible since they have nontoxic degradation products. Researchers use various combinations of isolated proteins or fully decellularized ECM. Decellularized ECM can be deposited by a chosen cell type in vitro or extracted from tissue samples or intact organ sections [1,9,17,[63][64][65][66]73].
Decellularized ECM provides a scaff old that preserves the complex interactions of the numerous ECM components, which is diffi cult to mimic with polymer scaff olds [63][64][65][66]. Gilbert describes methods and diff erence in tissue and organ decellularization [65]. However, decellularized ECM scaff olds can present problems of immunogenicity, as it is hard to achieve complete decellularization. Cellular and tissue debris can be left over, allowing foreign material to initiate an immune response. Specifi cally for vascular regeneration, Koffl er and colleagues used a biodegradable, acellular, Surgisis scaff old derived from porcine jejunum to create and study the integration of a vascularized muscle graft [73]. Part of the porcine small intestinal submucosa was taken from a pig and decellularized to create a small intestinal submucosa ECM-derived scaff old. Th e scaff old allowed for extended in vitro cell culture, vascularization, and muscle tissue organization, which resulted in improved anastomosis and vessel integration upon implantation. Overall, decellularization can provide an excellent approach for the generation of scaff olds as it preserves the physiological architecture, composition, and mechanics, which would support the formation of vasculature in vitro or the infi ltration of vasculature to repopulate the scaff old in vivo [63][64][65][66]. However, there are still challenges that need to be addressed in tissue engineering, such as the source of organs for human usage, obtaining enough cells to repopulate the decellularized matrix, and maintaining cell viability and continued function.
Collagens, specifi cally collagen type I, are commonly isolated to create an ECM protein-derived gel. Stratman and colleagues created a platform using a collagen type I matrix to explore the role of cytokines and growth factors in tube morphogenesis and sprouting [25]. Using the collagen scaff old, Stratman and colleagues found that VEGF and FGF prime ECs to respond to stem cell factor, IL-3, and stromal-derived factor-1α in serum-free conditions. Using this platform, these three cytokines were found to regulate EC morphogenesis and sprouting. Th is observation has major implications on current studies and clinical therapies, which apply pro-angiogenic factors. In a diff erent study by Au and colleagues, EPCs were found to form dense and durable vessels with 10T1/2 supporting cells in collagen-fi bronectin gels [74]. Another ECM-derived component used to study angiogenesis is HA, a glyco saminoglycan. We used a modifi ed HA hydrogel scaff old as a model for vascular network formation from human EPCs [62]. Vacuole and lumen formation, as well as branching and sprouting, were dependent on cell inter actions with RGD peptides presented on the HA scaff old. Hanjaya-Putra and colleagues observed anastomosis with the murine host circulatory system in vivo, creating a controlled tube morphogenesis model in a completely synthetic HA scaff old.
Signifi cant progress is being made with many scaff old materials in vascular engineering to promote and study vascular formation. Synthetic polymers provide high reproducibility and control over multiple parameters, allowing materials to be tuned for tissue-specifi c applications in the body. Natural polymers provide improved physiologic mimicry due to their biologically recogni zable side groups and biocompatible properties. Decellularized ECM scaff olds give researchers the advantage of using organization and composition that naturally occur in the body, especially with the preservation of threedimensional architecture. Current biodegradable scaff old platforms have increased the understanding of vascular network formation and the key signaling pathways involved. Th ese platforms have been mostly studied and assessed in vitro and on relatively small scales. To achieve a reproducible and reliable organ replacement therapy or ischemic tissue treatment, a deeper understanding of vascular functionality and durability in vivo needs to be explored. Altogether, platforms need to move from individual in vitro and small-scale animal trials to large animal models and human clinical studies in order to achieve pre-vascularize scaff olds and vascularization therapy of signifi cant clinical relevance.

Conclusion
Th ere is a signifi cant clinical need to engineer platforms that can promote angiogenesis in damaged, ischemic tissue or can regulate angiogenesis in cases of vascular overgrowth. Tissue engineering has increased our understanding of processes in vascular network formation. Currently, biodegradable scaff olds created from synthetic or natural polymers and ECM-derived scaff olds hold Figure 2. Example of a biodegradable scaff old platform to promote endogenous angiogenesis. Schematic of a dextran-polyethylene glycol diacrylate (PEGDA), three-dimensional, hydrogel scaff old promoting neovascularization, angiogenesis, and skin regeneration at a burn wound site. Reproduced with permission from Sun and colleagues [72]. promise in vitro and in animal studies. In many cases, however, scaff olds alone may not be enough to enable suffi cient recruitment of host vasculature to support tissue regeneration in a clinically relevant manner. Th ere is an increasing eff ort to understand the factors that control stem and progenitor cell homing and diff erentiation to vascular cell types, as well as the organization into vascular networks. One important aspect in the regulation of these processes is the physical interactions of cells with the scaff old prior to and after implantation. Presently, a quick off -the-shelf therapy to vascularize damaged tissue for any type of patient has yet to be achieved. Platforms need to be studied in preclinical, large animal models over extended time periods to truly gauge their clinical feasibility.