The Tissue Engineering Research Lab at California Polytechnic State University, San Luis Obispo focuses on creating tissue-engineered blood vessel mimics (BVMs) for use in preclinical testing of vascular devices. These BVMs are composed of electrospun scaffolds made of an assortment of polymers that are seeded with different cell types. This integration of polymers with cells leads to the need for biocompatibility testing of the polymer scaffolds. Many of the lab’s newest scaffolds have not been fully characterized for biologic interactions. Therefore, the first aim of this thesis developed methods for in vitro cytotoxicity testing of polymers used in the fabrication of BVMs. This included cytotoxicity testing using direct contact and elution-based methods, along with fluorescent staining to visualize the scaffold effects on cells. The second aim of this thesis implemented the newly developed cytotoxicity protocols to evaluate the biocompatibility of existing polymers, ePTFE and PLGA, used in the tissue engineering lab. The results demonstrated that ePTFE and PLGA were noncytotoxic to cells. The third aim of this thesis evaluated the biocompatibility of novel polymers used to fabricate BVMs: PLGA with salt, PLLA, and PCL. Elution-based methods concluded that PLGA with salt, PLLA, and PCL were noncytotoxic to cells; however, the direct contact method illustrated PLGA with salt and PCL were mildly cytotoxic at 24 and 48 hours. Potential causes of this variability include the addition of salt to PLGA, dissolving PCL in dichloromethane, inadequate sample sizing, and the inherent differences between the test methods. Overall, this thesis developed and implemented methods to evaluate the biocompatibility of polymer scaffolds used in the BVM model, and found that ePTFE, PLGA, and PLLA scaffold materials were biocompatible and could be implemented in future BVM setups without concerns. Meanwhile, PLGA with salt and PCL’s toxicity was mild enough to urge future cytotoxicity testing on PLGA with salt and PCL before further use in the lab.

“Blood vessel mimics” (BVMs) are tissue-engineered constructs that serve as in vitro preclinical testing models for intravascular devices. The Cal Poly Tissue Engineering lab specifically uses BVMs to test the cellular response to stent implantation. PLGA scaffolds are electrospun in-house using the current “Standard Protocol” and used as the framework for these constructs. The performance of BVMs greatly depends on material and mechanical properties of the scaffolds. It is desirable to create BVMs with reproducible properties so that they can be consistent models that ultimately generate more reliable results for intravascular device testing. Reproducibility stems from the consistency of the scaffolds. Thus, scaffolds with consistent material and mechanical properties are necessary for creating reproducible BVMs. The aim of this thesis was to characterize the reproducibility of the electrospun PLGA scaffolds using fiber diameter measurements and compliance testing. Initial work in this investigation involved designing and testing several experimental electrospinning protocols to obtain smaller fiber diameters, which have been shown to elicit more ideal cellular responses. The most successful protocol in that regard was then analyzed for the reproducibility of fiber diameters and compared to the reproducibility of the Standard Protocol. After determining that the Standard Protocol produced scaffolds with more consistent fibers, a large-scale reproducibility study was performed using this protocol. In this expanded study, both fiber diameter and compliance were analyzed and used to characterize the scaffolds. It was established that the scaffolds demonstrated inconsistent mean fiber diameter and mean compliance. The current standard electrospinning protocol therefore does not create PLGA scaffolds with statistically reproducible properties. Future modifications should be made to the electrospinning parameters in order to reduce variability between the scaffolds and future studies should be performed to determine the acceptable range of properties.

Optical coherence tomography (OCT) is a technology that enables 2D cross-sectional images of tissue microstructure. This interferometric technique provides resolutions of approximately 10-20 um with a penetration depth of 1-2 mm in highly scattering tissues. With the use of fiber optics, OCT systems have been developed for intravascular imaging with a demonstrated improvement in both resolution and dynamic range compared to commercial intravascular ultrasound systems. OCT studies of normal, atherosclerotic, and stented arteries indicate the ability of OCT to visualize arterial structures. These results suggest OCT may be a valuable tool for studying luminal structures in tissue engineered constructs.In the present study, new endoscopic OCT systems and analysis techniques were developed to visualize the growth and response of the cellular lining within a tissue engineered blood vessel mimic (BVM). The BVM consists of two primary components. A biocompatible polymeric scaffold is used to form the tubular structure. Human microvessel cells from adipose tissue are sodded on to the inner surface of the scaffold. These constructs are then developed and imaged within a sterile bioreactor.Three specific aims were defined for the present study. First, an OCT longitudinal scanning endoscope was developed. With this endoscope, a study of 16 BVMs was performed comparing images from OCT and corresponding histological sections. The study demonstrated that endoscopic imaging did not visually damage the mimic cellular lining. OCT images showed excellent correlation with corresponding histologicalsections. Second, a concentric three element endoscope was developed to provide radial cross-sections of the BVM. OCT images using this endoscope monitored lining development on three types of polymeric scaffolds. In the third specific aim, automated algorithms were developed to assess the percent cellular coverage of a stent using volumetric OCT images.The results of the present study suggest that OCT endoscopic systems may be a valuable tool for assessing and optimizing the development of tissue engineered constructs. Conversely, the BVMs modeled the arterial response to deployed stents allowing the development of automated OCT analysis software. These results suggest that blood vessel mimics may be used to advance OCT technology and techniques.

The work of the Cal Poly Tissue Engineering Lab is primarily focused on the fabrication, characterization, and improvement of “Blood Vessel Mimics” (BVMs), tissue engineered constructs used to evaluate cellular response to vascular medical devices. Currently, cells are grown onto fibrous, porous tubes made using an in-house electrospinning process from PLGA, a biocompatible co-polymer. The adhesion and proliferation of cells in a BVM is reliant on the micro-scale structure of the PLGA scaffold, and as such it is of great importance for the electrospinning process to consistently produce scaffolds of similar morphologies. Additionally, it has been shown that cell proliferation increases with scaffolds of smaller fibers and pores than the current electrospinning protocol can produce. Finally, the Tissue Engineering Lab has interest in testing devices in more tortuous BVM bioreactor designs, however the use of relatively rigid PLGA scaffolds has severely limited the ability to construct more complicated vessel geometries. The overall goal of this thesis was to improve fabrication and characterization of electrospun polymer scaffolds for BVM use. The specific aims of this thesis were to: 1) Improve scaffold characterization by comparing two techniques for fiber diameter measurement and implementing a technique for pore area measurement. 2) Reduce scaffold fiber diameter and pore area by investigating humidity and solvent composition electrospinning parameters. 3) Reduce process variability by developing a more specific electrospinning protocol. 4) Improve scaffold consistency and use by understanding and reducing PLGA scaffold shrinkage. 5) Identify and evaluate more flexible polymers as potential alternatives for electrospun BVM scaffolds. In order to accomplish these aims, first, several BVM and outside literature images were taken and evaluated with current and prospective fiber diameter techniques, and with 2 prospective pore area techniques to characterize accuracy and consistency of each method. It was found that the prospective fiber diameter measurement technique was not superior to the current method. The techniques developed for pore area measurement were found to produce results that differed significantly from each other and from the published value for a given image. Next, changes to environmental and solution composition parameters were made with the hopes of reducing fiber diameter and pore area of electrospun PLGA scaffolds. Changes in relative humidity did not appear to significantly affect scaffold fiber diameter while changes to solvent composition, specifically the use of acetone, resulted in fibers significantly smaller than those regularly achieved in the BVM lab. Next, several sources of variability in the electrospinning protocol were identified and subsequently altered to improve consistency and usability. Specifically, this included redefining the precision with which PLGA mass was measured, repositioning electrical equipment to reduce the effect of stray electrostatic forces on the polymer solution jet, attempting to control the temperature and humidity inside the electrospinning enclosure, and improving the ease with which scaffolds are removed from their mandrels through alternative mandrel surface treatments. In addition to overall process variability, the issue of scaffold shrinkage during BVM use was investigated and two possible treatments, exposure to either ethanol or elevated temperatures, were proposed based on previous electrospinning literature results. Each was tested for their effectiveness in mitigating shrinkage through exposure to BVM setup-mimicking conditions. It was found that both treatments reduced scaffold shrinkage compared to control samples when exposed to BVM setup-mimicking conditions. Finally, 3 flexible polymers were selected and electrospun to compare against typical PLGA results and to conduct a kink radius test as a metric for measuring flexibility as it pertains to the proposed BVM lab application. It was concluded that two types of thermoplastic polyurethane (tPU) were not acceptable electrospinning materials for use in the BVM lab. Additionally, while polycaprolactone (PCL) could be successfully electrospun it could not undergo the amount bending required for more tortuous BVM bioreactor designs without kinking. Overall, the work in this thesis provided insight into multiple scaffold characterization techniques, reduced overall electrospinning variability in the fabrication and use of PLGA scaffolds, and defined processing parameters that have been shown to yield scaffolds with smaller morphological features than all prior Tissue Engineering Lab work. By creating better, more effective scaffolds, researchers in the Tissue Engineering Lab can more accurately mimic the structure and properties of native blood vessels; this, in turn, will result in BVM cell responses that more closely resemble that of native tissue. Creating consistent and appropriate BVMs will then lead to impactful contributions to the existing body of tissue engineering research and to better preclinical device testing.

Tissue engineering is an emerging field that offers novel and unmatched potential medical therapies and treatments. While the vast aim of tissue engineering endeavors is to provide clinically implantable constructs, secondary applications have been developed to utilize tissue-engineered constructs for in-vitro evaluation of devices and therapies. Specifically, in-vitro blood vessel mimics (BVM) have been developed to create a bench-top blood vessel model using human cells that can be used to test and evaluate vascular disease treatments and intravascular devices. Previous BVM work has used fat derived human microvascular endothelial cells (EC) sodded on an ePTFE scaffold. To create a more physiologically accurate model, a dual layer of large vessel endothelial and smooth muscle cells (SMC) on an ePTFE tube is investigated throughout this thesis. Human umbilical vein endothelial cells (HUVEC) and human umbilical vein smooth muscle cells (HUVSMC) were chosen as the large vessel cell types and cultivated according to standard procedures. Before dual sodding, sodding density experiments with HUVSMC were performed to determine the number of cells required to create a confluent cell layer. HUVSMC sodded by trans-luminal pressure at densities ranging from 3.5x10^5 cells/cm^2 to 1.0x10^6 cells/cm^2 were run for one day to observe luminal coverage. After determining the desirable range for HUVSMC sodding, HUVSMC experiments with 5.0x10^5 cells/cm^2 and 7.5x10^5 cells/cm^2 were run over seven days to evaluate progression of the graft over time. Histology and SEM methods were used for analysis. A HUVEC study was next conducted over 7 days to confirm that the large vessel endothelial cell could be sodded and sustained on ePTFE in-vitro. Next, dual sodding was performed by pressure sodding HUVSMC at 7.5x10^5 cells/cm^2 followed by trans-luminal flow for 30 minutes. HUVECs were subsequently trans-luminally pressure sodded at 5.0x10^5 cells/cm^2 followed by an additional 30 minutes of trans-luminal flow; perfusion flow began following the final 30 minutes of trans-luminal flow. Experiments for the dual layered grafts were run for both one and seven days to evaluate and develop the dual sodding protocol as well as observe the co-culture over time. Analysis of the dual layered grafts was performed by SEM, histology, and fluorescence microscopy. HUVECs were incubated with Cell Tracker™ prior to dual sodding and both cell types with bisbenzimide after graft harvest to attempt to distinguish between cell types. Results from the thesis illustrate that large vessel smooth muscle and endothelial cells can be sodded onto ePTFE scaffolds and sustained within the in-vitro BVM system for up to 7 days. Furthermore, cost analysis demonstrates that the addition of a smooth muscle cell layer adds minimal costs to the BVM system. In conclusion, the studies contained within this thesis culminate in a protocol for the dual sodding of smooth muscle and endothelial cells with the aim of creating a physiologically representative co-culture blood vessel mimic.

Every 34 seconds, someone in the United States suffers from a heart attack. Most heart attacks are caused by atherosclerotic build up in the coronary arteries, occluding normal blood flow. Balloon angioplasty procedures in combination with a metal stent often result in successful restoration of normal blood flow. However, bare metal stents often lead to restenosis and other complications. To compensate for this problem, industry has created drug-eluting stents to promote healing of the artery wall post stenting. These stents are continually advancing toward better drug-eluting designs and methods, resulting in a need for fast and reliable pre-clinical testing modalities. Dr. Kristen Cardinal recently developed a tissue engineered blood vessel mimic, with the goal of testing intravascular devices. However, the scaffold component of this model exhibits several physiological limitations that must be addressed to create a truly biomemtic BVM. The current model uses expanded poly(terafluorethylene) [ePTFE] or poly(lactic-go-glycolide) [PLGA] as the choice material for the scaffold. EPTFE has several advantages as it is a widely recognized biomaterial. However, ePTFE is very expensive and lacks native mechanical properties. PLGA is another polymer that is created in-house to produce a uniquely tailored scaffold for use in the BVM; resulting in a cheaper alternative scaffold material. However, PLGA again lacks the necessary native mechanical properties to properly mimic an in-vivo artery. The creation of a biological scaffold will provide a unique biomimetic material to most accurately recapitulate the artery in-vitro. Decellularization is the process of removing all cellular components from a tissue, leaving an acellular structure of extracellular matrix. Understanding the clinical problem and the potential of the BVM, the aim of this thesis is to develop the decellularization process for the creation of a biologic scaffold as a replacement to the non-physiologic polymer scaffolds for the BVM. The first phase of this thesis was to develop and optimize an acceptable protocol for the decellularization of porcine arteries. The use of a 0.075% sodium dodecyl sulfate detergent was sufficient for complete removal of all vascular cell types, without significant degradation to the scaffold wall. In the second phase of this thesis, the decellularized scaffolds were mechanically tested to ensure retention of their native properties. The longitudinal and radial properties of the scaffold were found to be similar to the native artery, indicating the decellularized scaffold improves several physiologically aspects when compared to a polymer scaffold. These mechanical attributes improve the testing environment when evaluating sent deployment or new balloon angioplasty devices; as the decellularized scaffold has an phsyiolgical compliance. The final phase of this thesis examined the cellular adhesion capacities of the scaffold through recellularization with human umbilical vein endothelial cells (hUVECS). Fluorescent microscopy analysis suggests uniform attachment of cells along the length of the scaffold creating a monolayer. These results indicate this new scaffold type can develop an endothelium to complete the ideal, most physiologically relevant BVM system. Further optimization of the decellularization procedures could enhance the ability of the scaffold to be cultured for long-term interaction with intravascular devices.

The prevalence and devastating nature of cardiovascular diseases has led to many advancements in the therapies used to treat the millions of patients that suffer as a result of these conditions. As coronary artery disease (CAD) is the most common of these cardiovascular conditions, it is a major focus of research among the medical industry. Although lifestyle changes and drug therapies can treat early CAD, more advanced cases often require more definitive interventions. In conjunction with angioplasty, stenting of an occluded vessel has shown significant success in preventing restenosis. However, as with nearly every therapeutic process in the medical field, several complications have arisen in stented patients that pose a need for further improvement of the devices. As a result, the stent industry is constantly striving towards improving the characteristics and outcome of their product and with these efforts comes the need for extensive testing and research. Continuous improvement and innovation in the field of tissue engineering has brought about the possibility of creating laboratory grown tissue engineered vascular grafts (TEVGs) for the purpose of replacing and/or bypassing damaged or occluded regions of the vasculature. By employing the techniques used to produce TEVGs, a blood vessel mimic (BVM) bioreactor system has been developed with the intent of using the resulting construct as a model for testing the cellular response of a human blood vessel to an intravascular device such as a stent. This would allow gathering of more significant data in the early stages of device development and may reduce the overall costs and time required to refine a design. Although the BVM system has previously been used to cultivate viable constructs that were subsequently used to observe the response to a deployed stent, the flow conditions within the original design are not representative of the physiologic conditions in a native vessel. This aspect of the original system presented a need for development in order to be considered by researchers as an accurate in vitro representation of the target vessels in which the stents are used. One of the primary concerns of this environment is creating and maintaining physiologic flow conditions that will represent those present in native vessels in order to facilitate cells sodded on the construct to grow as they would under native conditions. The two key aspects of flow are pulsatility and wall shear stress. Studies in this thesis were carried out to determine the best and most feasible methods for implementing appropriate levels of pulsation and wall shear stress in the previously established BVM bioreactor system with the intention of maintaining the original system’s simplicity and high throughput potential. Pulsatile flow was created by elevating backpressure in the BVM chamber while using a different pump head and pump tubing. Wall shear stress was adjusted by altering the viscosity of the perfusate and flow rate through the system. Both pulsatile flow and shear stress were established without any major changes to the overall configuration of the system. Pulsatile pressures of ~80 mmHg and wall shear stress forces of ~6.4 dyn/cm2 were established with minimal alteration to the original system. Pulsatility was created by using a 3-roller peristaltic pump head in place of the originally specified 8-roller head to create pulses that were then regulated with backpressure created by restricting down stream flow. Increasing the viscosity and corresponding flow rate allowed for instigation and control of wall shear stress at the inner wall of the BVM graft. Although the resulting protocols presented here require refinement for ultimately successful implementation, they are important underpinnings that will facilitate the eventual development of an ideal BVM system that is highly suitable for use as a high-throughput intravascular device testing model.

Introduction: Diabetes Mellitus is a metabolic disorder that affects a person’s ability to either produce insulin (Type I diabetes mellitus) or properly use insulin (Type II diabetes mellitus) in order to maintain adequate blood glucose levels. The most severe diabetic complications arise due to hyperglycemia – a state of extremely high blood glucose levels – such as, coronary artery disease (CAD), in which coronary stent therapy is a popular method of treatment. However, research has shown a high rate of in-stent restenosis in diabetic patients with CAD, most likely due to activation of cellular adhesion molecules on endothelial cells exposed to the hyperglycemic environment. Blood vessel mimics (BVMs) have been researched as viable options for in vitro studies on vascular stents; thus, it would be beneficial to create an in vitro diabetic BVM for stent manufactures to evaluate and determine the root cause of the high failure rate of stents in the diabetic population. In addition, a diabetic BVM would help manufactures optimize coatings or stent configurations for diabetic patients. Methods: The purpose of this thesis was to take the initial steps towards the goal of a diabetic BVM. The first aim was to establish a procedure of developing glycemic cell media solutions of various glucose concentrations, and to establish a feasible method of monitoring the glucose concentration of the solutions. Glycemic cell media solutions were developed and their glucose concentrations were evaluated with a blood glucose meter (specifically, the Aviva Accu-Chek blood glucose meter) or visual blood glucose test strips (Glucoflex R visual blood glucose test strips). The second aim was to ensure that the developed glycemic cell media solutions could be monitored in a cell culture environment over time, and to determine if the hyperglycemic conditions induced any change to endothelial cells. Bovine aortic endothelial cells (BAECs) and human umbilical vein endothelial cells (HUVECs) were used to evaluate glucose consumption and cell morphology. Glucose concentration of the cell media was recorded to evaluate glucose consumption, and the cells were evaluated under a microscope in order to determine cell morphology and an increase in cell death. Results & Conclusions: Data accumulated from the first set of experiments confirmed that glycemic cell media solutions can be developed by adding Sigma G6512 D-(+)-glucose to base cell media. Additionally, the Aviva Accu-Chek blood glucose meter recorded the most accurate and precise glucose concentrations of the various glycemic cell media solutions compared to the Glucoflex-R blood glucose visual test strips. Lastly, the series of experiments with BAECs and HUVECs confirmed that the glycemic cell media solutions could be effectively monitored over time, and that these conditions evoked higher glucose consumption by the endothelial cells compared to the normal glycemic cell media solutions. Additionally, neither glycemic environment evoked significant cell death. These results met the aims of this thesis, and therefore provide the foundation for further development of a diabetic BVM.

Intravascular devices, such as stents, must be rigorously tested before they can be approved by the FDA. This includes bench top in vitro testing to determine biocompatibility, and animal model testing to ensure safety and efficacy. As an intermediate step, a blood vessel mimic (BVM) testing method has been developed that mimics the three dimensional structure of blood vessels using a perfusion bioreactor system, human derived endothelial cells, and a biocompatible polymer scaffold used to support growth of the blood vessel cells. The focus of this thesis was to find an in-house fabrication method capable of making cellular scaffolding for use in the BVM. Research was conducted based on three aims. The first aim was to survey possible fabrication methods to choose a technique most appropriate for producing BVM scaffolding. The second aim was to set up the selected fabrication method (electrospinning) in-house at Cal Poly and gain understanding of the process. The third aim was to evaluate consistency of the technique. The work described in this thesis determined that electrospinning is a viable fabrication technique for producing scaffolding for BVM use. Electrospun scaffolding is highly tailorable, and a structure that mimics the natural organization of nano sized collagen fibers is especially desirable when culturing endothelial cells. An electrospinning apparatus was constructed in house and a series of trial experiments was conducted to better understand the electrospinning process. A consistency study evaluated scaffold reproducibility between different spins and within individual spins while setting a baseline that can be used for comparison in future work aimed at electrospinning.

Coronary artery disease affects millions of people and the ability to detect and treat the disease is advancing at a rapid rate. As a result, the development of intravascular technologies is the focus of many medical device manufacturers. Specifically, coronary stent implantation is being performed in an increasing number of patients and a number of new stent designs have been introduced to the market, resulting in the need for improved preclinical testing methods. An in vitro tissue engineered “blood vessel mimic” (BVM) system has previously been established and its feasibility for the initial testing of newly emerging intravascular technology has been demonstrated. There are limitations that exist with this original design, however, and the focus of this thesis was to both improve and expand upon the original model. Therefore, research was conducted based on two specific aims. The first aim was to develop a more ideal BVM system to accommodate a wider range of stent lengths and diameters, while allowing for easy graft insertion and seal-ability. The second aim was to develop next generation BVM systems,focused on future needs and technology, such as long, angulated and bifurcated geometries. The work described in this thesis demonstrates that a BVM chamber can be created which has the advantages of easy graft insertion and seal-ability, as well as the ability to accommodate varying sizes of vessel scaffolds, all while maintaining the needs of a tissue engineering bioreactor system. The next generation BVM systems presented demonstrate that the BVM concept can be expanded to meet the needs of long, angulated and bifurcated geometries. Overall, the work in this thesis describes the design and optimization of an in vitro blood vessel mimic bioreactor system for the evaluation of intravascular devices, specifically coronary stents, in simple and complex vessel geometries.