Small diameter vascular grafts comprised entirely from cells and cell-derived extracellular matrix (ECM) have shown promise in clinical trials and may have potential advantages as in vitro vascular tissue models. A challenge with current cell-derived tissue engineering approaches is the length of time required to generate strong, robust tissue. There is a lack of alternative methods to rapidly assemble cells into a 3D format without the support of a scaffold. Toward the goal of engineering a new approach to rapidly synthesizing vascular tissue constructs entirely from cells, we have developed and characterized a strategy for creating cell-derived tissue rings by cellular self-assembly. The focus of this thesis was to develop the system to rapidly generate engineered tissue rings, and to evaluate their structural and functional properties. To generate tissue rings, rat smooth muscle cells (SMCs) were seeded into round-bottomed, ring-shaped agarose wells with varying inner post diameters (2, 4, and 6 mm). Within 24 hours of seeding, cells aggregated, contracted, and formed robust tissue that could be removed from their wells and handled. If kept in culture, the thickness of these tissue rings increased with time. Mechanical analysis of the tissue showed that it was stronger after only 8 days in culture than engineered tissues generated by other approaches (such as seeding cells in biopolymer gels) cultured and tested at similar time points. Histological staining of the tissue rings revealed high cell densities throughout, along with the presence of glycosaminoglycans and some collagen. We also found that we could use the tissue rings as building blocks to generate larger tubular structures. Briefly, tissue rings were removed from the agarose wells and transferred onto silicone tubing mandrels. Once the rings were placed in contact with each other on the mandrel, they were cultured to allow the rings to fuse together. We found that the ability of tissue rings to fuse decreased with increasing ring “pre-culture� duration, and that we were able to generate fully fused tissue tubes in as little as 8 days (with only one day of ring pre-culture and seven days of fusion). In the last section of this thesis, we established the feasibility of using primary human SMCs to generate self-assembled tissue rings, similar to the self-assembled rings generated with rat SMCs. Compared to the rat SMC rings, human SMC rings were stronger, stiffer and appeared to contain increased levels of collagen. These data showed that human SMCs are capable of self-assembling into tissue rings similar to rat SMCs, and may therefore be used to create engineered human vascular tissue. Overall, we have developed a platform technology that can be used to screen the effects of culture parameters on the structure, mechanics, and function of vascular tissue. We anticipate that through the use of this technology, we can further improve vascular grafts by better understanding factors which promote ECM synthesis and SMC contraction. We can use these results directly toward the generation of vascular grafts by fusing self-assembled cell rings together to form tissue tubes. These novel bioengineered vascular tissues may also serve as a method to produce in vitro models to help further our understanding of vascular diseases, as well as facilitate pre-clinical screening of vascular tissue responses to pharmacologic therapies.

Key Words: Thrombosis, Vascular Graft, Microfluidics, Wall Shear Stress / Vascular grafts are essential for the management of cardiovascular disease. However, the lifesaving potential of these devices is undermined by thrombosis arising from material and flow interactions on the blood contacting surface. To combat this issue, the use of antithrombogenic coatings has emerged as a promising strategy for modulating blood and graft interaction in vivo. Although an important determinant of graft performance, hemodynamics are frequently overlooked in the in vitro testing of coatings and their translatability remains poorly understood. We address this limitation with a microscale platform that incorporates vascular prosthesis and coatings with tuneable flow and surface conditions in vitro. As a proof of concept, we use the platform to test the thrombogenic performance of a novel class of lubricant infused (LIS) and antibody lubricant infused (anti-CD34 LIS) coated ePTFE vascular grafts in the presence of arterial wall shear stress, with and without the presence of endothelial cells. Our findings suggest lubricant infused coated ePTFE vascular grafts are thromboresistant under flow and may have potential for in vivo arterial grafting applications. It is moreover apparent that the microscale properties of the device could be advantageous for the testing and translation of novel antithrombogenic coatings or blood contacting prosthesis in general. / Thesis / Master of Applied Science (MASc)

Abdominal aortic aneurysm (AAA) is a prevalent disease in developed countries. While endovascular aneurysm repair is fairly successful, it has shortcomings. Polymeric endoluminal paving and sealing is a method that has previously been developed to treat a range of diseases. Our goal is to further develop this technique to treat AAA, a process we have named polymeric endo-aortic paving (PEAP). We hypothesize that PEAP will overcome many of the limitations associated with EVAR by providing a minimally invasive treatment which can be used on patients with complicated AAA geometries and reducing incidence of migration and endoleak. Additionally, we plan to incorporate drug delivery into PEAP to improve efficacy. The purpose of this work was to evaluate a potential graft material for PEAP and to develop a thrombus mimic which will aid in further PEAP development. Blends of polycaprolactone/polyurethane (PCL/PU) were assessed by characterizing their mechanical, thermoforming, and degradation properties. PCL/PU grafts have a similar stiffness to aortic tissue and can be thermoformed at temperatures approaching 37 degrees C. Blending PCL with PU significantly reduces PCL's degradation. An anisotropic hyperelastic strain energy function was developed for the PCL/PU blends and finite element modeling (FEM) was used to show that stress reduction on the AAA wall that can be achieved by PEAP is similar to current EVAR. Stiffness varies throughout the AAA thrombus, and thrombus mimics were developed that have similar stiffness, components, and structure to native AAA thrombus.

Electrospinning is a polymer processing technique which allows for the production of nano to micro size fibers and scaffolds which can be composed of numerous synthetic biodegradable materials and natural biopolymers. Natively, elastin and collagen are the main components of vascular tissue. Arranged in a tri-layered structure, they create a specific mechanical environment that can withstand the rigors of circulation. The goal of this study was to develop a mechanically ‘biomimicking’ vascular graft composed of three distinct layers through the process of electrospinning. We hypothesize that the use of bioactive agents such as elastin, collagen, and silk to supplement poly(caprolactone) at specified ratios for each layer would provide a finely tuned vascular replacement. This was accomplished by establishing cross-linking parameters for the biopolymer materials and then assessing the mechanical properties of individual materials and eventually a whole tri-layered graft. Additionally, while mechanical testing can lead to a good graft, a replacement graft requires excellent cellular properties as well to promote cell infiltration, proliferation, and migration. Therefore, the conclusion of this study examines the integrin binding characteristics of the electrospun biopolymers. First, the results from the preliminary cross-linking study examined the dissipation of soluble elastin when uncross-linked v. cross-linked. It was determined through this initial study that synthetic scaffolds blended with soluble proteins such as elastin require a fixation in order to retain their protein mass within the scaffold. Retaining this mass, incrementally changed the material properties of the blended scaffolds. This initial study was then carried further to establish optimal cross-linking parameters using two different types of reagents: carbodiimide and genipin. It was found that lower cross-linking molarities produced excellent results based on assays performed to assess cross-linking percentages and rate of reaction. Some differences in mechanical properties were seen, but they did not constitute a choice of one cross-linker over the other. The next portion of this study aimed to design a tri-layered graft. This was performed with the aid of mathematical analysis to observe circumferential wall stresses based on simple tensile properties. A series of tri-layered grafts were electrospun using poly(caprolactone), elastin, and collagen. The medial layers of these grafts were changed while the intima and adventitia remained constant. Differences were demonstrated as the elastin content of the medial layer decreased, proving that each layer had an affect on the overall graft properties and that it was possible to tune graft mechanics. A larger tri-layered study looked to evaluate changes in the adventitial and medial layers while keeping the intimal layer constant using poly(caprolactone), elastin, collagen, and silk fibroin. In this study, differences were exhibited under compliance and burst strength testing, narrowing the scope of material choices. Results from a 4 week degradation study with the best tri-layered grafts revealed no evidence of degradation, but did generate some positive compliance results for two of the grafts. Finally, integrin binding and protein analysis portrayed results that were indicative of the existence of ligand binding sites for collagen scaffolds and the possibility of a small amount of ligand sites on silk. Elastin, however, displayed low to non-existent adhesion. These studies produced results that allowed us to continuously narrow the scope of materials as the experiment progressed towards an optimized tri-layered vascular graft.

Current synthetic vascular grafts have high failure rates in small-diameter (<6 mm) applications due to inadequate cell-material interactions and poor matching of arterial biomechanical properties. To address this, we have developed a multilayer vascular graft design with a non-thrombogenic inner layer that promotes endothelial cell (EC) interactions and a reinforcing layer with tunable biomechanical properties. The blood-contacting layer of the graft is based on a Streptococcal collagen-like protein (Scl2-1). Scl2-1 has the triple helical structure of collagen, but it is a non-thrombogenic protein that can be modified to have selective cell adhesion. For this application, Scl2-2 has been modified from Scl2-1 to contain integrin binding sites that promote EC adhesion. We have developed the methodology to incorporate Scl2 proteins into a poly(ethylene glycol) (PEG) hydrogel matrix. PEG-Scl2 hydrogels facilitate optimization of both bioactivity and substrate modulus to offer unique control over graft endothelialization. However, scaffold properties that promote endothelialization may not be consistent with the mechanical properties necessary to withstand physiological loading. To address this issue, we have reinforced PEG-Scl2-2 hydrogels with an electrospun polyurethane mesh. This multilayer vascular graft design decouples requisite mechanical properties from endothelialization processes and permits optimization of both design goals. We have confirmed the thromboresistance of PEG-Scl2-2 hydrogels in a series of whole blood tests in vitro as well as in a porcine carotid artery model. Additionally, we have shown that the electrospun mesh biomechanical properties can be tuned over a wide range to achieve comparable properties to current autologous grafts. Traditional acrylate-derivatized PEG (PEGDA) hydrogels were replaced with PEG diacrylamide hydrogels with similar properties to increase biostability for long-term implantation. These findings indicate that this multilayer design shows promise for vascular graft applications. As vascular graft endothelialization can significantly improve success rates, the ability to alter cell-material interactions through manipulations in PEG-Scl2-2 hydrogel properties was studied extensively. By reducing Scl2-2 functionalization density and utilizing a biostable PEG functionalization linker, Acrylamide-PEG-I, significantly improved initial EC adhesion was achieved that was maintained over 6 weeks of swelling in vitro. Additionally, increases in Scl2-2 concentration and in hydrogel modulus provided increased EC interactions. It was found that PEG-Scl2-2 hydrogels promoted enhanced EC proliferation over 1 week compared to PEG-collagen gels. In summary, we have developed a vascular graft with a biostable, non-thrombogenic intimal layer that promotes EC adhesion and migration while providing biomechanical properties comparable to current autologous grafts. This design demonstrates great potential as an off-the-shelf graft for small diameter arterial prostheses that improves upon current clinically available options.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22349号 / 医博第4590号 / 新制||医||1042(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 木村 剛, 教授 椛島 健治, 教授 妻木 範行 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Cardiovascular disease is currently the leading cause of death in the U.S. that frequently requires bypass surgery using vascular grafts for treatment. Current limitations with fully synthetic grafts have led researchers to bioengineered alternatives that consist of a combination of vascular scaffolds and cells. A major challenge in creating a functional bioengineered vascular graft is development of a confluent endothelium on the lumen that is able to resist detachment under physiologic fluid flow. In addition, methodologies used to assess the growth and maturation of the endothelium in a noninvasive and dynamic manner are severely lacking. Therefore, the overall goal of this research is to advance the field of vascular tissue engineering by 1) creating methodologies to enhance EC adherence to a vascular graft and 2) development of a noninvasive and real-time imaging system capable of assessing the graft endothelium.  To achieve these objectives, three separate studies were performed. In the first study, electrospun scaffold fiber diameter and alignment were systematically varied to determine their effect on endothelial cell (EC) morphology and adherence under fluid flow. ECs on uniaxially aligned nanofibers displayed elongated and aligned morphologies leading to higher adherence to the scaffolds under physiologic levels of fluid flow as compared to those on randomly oriented scaffolds. In the second study, a fiber optic based (FOB) imaging system was developed to image fluorescent ECs through a thick electrospun scaffold.  Results demonstrated that the FOB imaging system was able to accurately visualize fluorescent ECs in a noninvasive manner through the thick and highly opaque scaffold. In the final study, the FOB imaging system was used to noninvasively quantify vascular graft endothelialization, EC detachment, and apoptosis through the vessel wall with greater imaging penetration depth than two-photon microscopy. Additionally, the FOB method was capable of continuously tracking EC migration and endothelialization of a bioengineered graft in a bioreactor. Overall, these results demonstrate that aligned scaffold topographies enhance EC adherence under fluid flow and the FOB imaging system is a promising tool to monitor endothelium development and response to fluid flow in a manner that has not previously been afforded using conventional imaging methods. / Ph. D.

No description available.

The resurgence, and eventual rise to prominence in the field of tissue engineering, that electrospinning has experienced over the last decade speaks to the simplicity and adaptability of the process. Electrospinning has been used for the fabrication of tissue engineering scaffolds intended for use in nearly every part of the human body: blood vessel, cartilage, bone, skin, nerve, connective tissue, etc. Diverse as the aforementioned tissues are in both form and function, electrospinning has found a niche in the repair of each due to its capacity to consistently create non-woven structures of fibers ranging from nano-to-micron size in diameter. These structures have had success in tissue engineering applications because of their ability to mimic the body’s natural structural framework, the extracellular matrix. In this study we examine a number of different techniques for altering scaffold properties (i.e. mechanical strength, degradation rate, permeability, and bioactivity) to create electrospun structures tailored to unique tissue specific applications; the end goal being the creation of a cellularized tissue engineering ligament analogue. To alter the mechanical properties of electrospun structures while maintaining high levels of bioactivity, synthetic polymers such as polydioxanone were blended in solution with naturally occurring proteins like elastin and fibrinogen prior to electrospinning. Cross-linking of electrospun structures, using glutaraldehyde, carbodiimide hydrochloride, and genipin, was also investigated as a means to both improve the mechanical stability and slow the rate of degradation of the structures. Fiber orientation and scaffold anisotropy were controlled through varying fabrication parameters, and proved effective in altering the mechanical properties of the structures. Finally, major changes in the structure of electrospun scaffolds were achieved through the implementation of air-gap electrospinning. Scaffolds created through air-gap electrospinning exhibited higher porosity’s than their traditionally fabricated counterparts, allowing for greater cell penetration into the scaffold. Overall, this collection of results provides insight into the diversity of electrospinning and reveals innumerous options, both pre and post fabrication, for the tissue engineer to create site-specific engineering scaffolds capable of mimicking both the form and function of native tissue.

<p>Synthetic small diameter vascular grafts fail clinically due to thrombosis and intimal hyperplasia. The attachment of endothelial cells (ECs) onto the inner lumen of synthetic small diameter vascular grafts can improve graft patency; however, significant challenges remain that prevent wide clinical adoption. These issues include difficulties in the autologous sourcing of ECs, the lack of attachment, growth and retention of the layer of ECs to the graft lumen, and the maintenance of an anti-thrombotic and anti-inflammatory profile by the layer of ECs. </p><p>This dissertation describes the isolation, characterization, and use of endothelial progenitor cells (EPCs) to improve the performance of small diameter vascular grafts. First, EPC isolation efficiency and expression of critical EC markers was compared between young healthy volunteers and patients with documented coronary artery disease (CAD). EPCs were isolated and expanded from patients with CAD and had a similar phenotype to EPCs isolated from healthy donors, and a control population of human aortic ECs. Second, we assessed the ability to enhance the anti-thrombotic activity of patient derived EPCs through the over expression of thrombomodulin (TM). In vitro testing showed TM-transfected EPCs had significantly increased production of key anti-thrombotic molecules, reduced platelet adhesion, and extended clotting times over untransfected EPCs. Finally, native and TM-transfected EPCs were seeded onto small diameter vascular grafts and tested for their ability to improve graft performance. EPCs sodded onto the lumen of small diameter ePTFE vascular grafts had strong adhesion and remained adherent during graft clamping and exposure to flow. TM-transfected EPCs improved graft anti-thrombotic performance significantly over bare grafts and grafts seeded with native EPCs. Based on these promising in vitro results, grafts were implanted bilaterally into the femoral arteries of athymic rats. Bare grafts and grafts with air removed clotted and had only 25% patency at 7 days. In contrast, graft sodded with native EPCs or TM-transfected EPCs had 87% and 89% respective patency rates. High patency rates continued with 28 day implant testing with EPC sodded grafts (88% Native; 75% TM). There were no significant differences in patency rates at 7 or 28 days between native and TM-transfected grafts. These in vivo data suggest patient blood-derived EPCs can be used to improve the performance of small diameter vascular grafts.</p> / Dissertation