Mark M. Samet

A new method for continual quantitation of viable cells on endothelialized polyurethanes

Many of the segmented polyurethanes currently used in cardiovascular prostheses undergo either modification of their surface structure or are lined with a confluent monolayer of endothelial cells to improve their hemocompatibility. During the establishment of an endothelial cell lining on these biopolymers it is necessary to continually monitor the number of viable cells that are covering the substrate. Yet, not all of the conventional cell enumeration techniques are suitable for assessing the growth of endothelial cells on polyurethanes. Methods, such as direct cell counting, dye uptake, or DNA or protein staining require either a transparent scaffold or lead to termination of the culturing process prior to measurement. In addition, some of the spectroscopic assays are often hampered by interaction of the dyes and/or solubilizers with the various constituents (e.g., catalyzers, antioxidants) and/or functional groups in the polyurethane formulations. In addressing these problems, we adapted a novel, highly reproducible fluorescent assay which is based on reduction by viable cells of an electrochemically sensitive compound, Alamar Blue. The bioreduced product is soluble and stable in culture media and noncytotoxic. In addition, the assay is independent of the geometry or physicochemical properties of the polymeric surfaces. In the present study we focus on the implementation of this assay to monitoring attachment and growth of various endothelial cell types on segmented polyurethanes.

A new, cryoprecipitate based coating for improved endothelial cell attachment and growth on medical grade artificial surfaces.

Monoprotein coatings of biomaterials with either natural adhesion molecules or genetically designed analogs have been used to facilitate attachment and spreading of endothelial cells. However, such treatments were found insufficient to maintain the integrity of the endothelial surface under turbulent flow conditions. In addition, when brought into contact with blood, these coatings were susceptible to plasma and cell proteinases that could readily destroy their structure and weaken cell adherence to the surface. In addressing these problems, we developed a cryoprecipitate-based coating that can firmly bind to any nonporous, prosthetic surface and interact with endothelial cells. The primary structure of the coating consisted of an autologous fibrin meshwork. It was refined by various compositions of the fibrinogen containing mixture and secured to polystyrene or polyurethane surfaces by dry-heat treatment. Further modulation of the coating was achieved by physically immobilizing various doses of heparin and insulin into the three dimensional matrix of the meshwork. Endothelial cells attached and grew much better on polyurethanes coated with this autologous protein complex than on a polystyrene tissue culture surface. With proper use of its capacity to mimic the properties of basal membrane, and absence of immunologic complications, the resulting coating may become an ideal multifunctional interface between cells and prosthetic materials.

In Vitro Testing of Endothelial Cell Monolayers Under Dynamic Conditions Inside a Beating Ventricular Prosthesis

Thromboembolic complications remain a major problem associated with the long-term clinical use of cardiac prostheses. A promising approach toward resolving this predicament is lining the blood contacting surfaces with a functional monolayer of endothelial cells (EC). In developing an endothelialized cardiac prosthesis, the authors in the past focused on establishing a confluent EC monolayer on the luminal surface of ventricular blood sacs. In this study, the authors concentrated on exposing the post confluent monolayers to the dynamic conditions inside a beating ventricle. The cells, derived from either bovine aortae or jugular veins, were grown to post confluence inside fully assembled ventricles on fibronectin or plasma cryoprecipitate coated, textured surfaces. After 11 days of culturing under static conditions, the endothelialized ventricles were connected to a mock loop that was run for 6 and 24 hr at 60 bpm and mean flow rate of 3.2 L/min. The status of the monolayer was evaluated by Alamar Blue assay before and after each run, and the extent of surface coverage was determined visually using bright field microscopic study after cell staining with KMnO4 and toluidine blue. In addition, morphometric information on cells/polyurethane surface was obtained with a scanning electron microscope. After 6 hr of pumping, cell staining revealed signs of moderate cell loss in fibronectin coated blood sacs, whereas in cryoprecipitate coated bladders the signs of denudation were marginal. In seven ventricles operated for 24 hr, Alamar Blue measurements indicated 35 +/- 16% of cell loss from monolayers established on fibronectin coating, but only 4.8 +/- 6.25% on cryoprecipitate. Thus, the current study demonstrates the feasibility of maintaining an intact endothelial surface in a beating ventricular prosthesis and indicates that the integrity of the endothelial lining is dependent upon a proper choice of surface macrostructure and protein coating.

Effects of a total artificial heart right stroke volume limiter on left-right hemodynamic balance

The long-term maintenance of patients with failing hearts on cardiac prostheses requires prevention of device related thromboembolic events. This challenge is being addressed by endothelialization of the blood sacs. However, the practice of establishing and maintaining a durable endothelial cell monolayer inside a beating prosthesis has not been fully realized. Thus, before exposing endothelial cell monolayers to the hemodynamics inside an artificial heart, the authors studied the effect of various flow patterns in a ventricle shaped chamber on the integrity and morphology of the endothelium. After 20 hours of superfusion by pulsatile flow, there were no denudation signs in the jet, where shear stress was 1.5 dynes/cm2. However, there was measurable damage to the monolayer close to the periphery of the eddies (turbulent flow) at 0.15 dynes/cm2. In either case, there were no signs of cell alignment with the flow, but there were changes in cell morphology compared with that of static control. These findings suggest that adjustment of endothelial cells in response to frictional forces occurs even at low shear stresses and that random velocity fluctuations might jeopardize the integrity of endothelial cell monolayers.

Flow patterns and endothelial cell morphology in a simplified model of an artificial ventricle

The aim of this study was to delineate the flow patterns in a non-unidirectional flow field inside a ventricle-shaped cell culture chamber, and examine the resulting morphology and integrity of the endothelium in select regions of the monolayer. The chamber was perfused by pulsatile flow, and the coherent motion of the fluid was studied using flow visualization aided by image analysis. Four distinct flow patterns were discerned and examined: central jet, flow impingement, flow separation, and recirculating eddies. The influence of these patterns on endothelial cell morphology was assessed after 20 h of exposure to flow. There were no signs of damage to the endothelium in the jet region nor was there evidence of cell alignment with the flow. Yet, there were changes in cell morphology and cytoskeletal architecture as compared to control. By contrast, within the eddies where the flow was highly disturbed, there was apparent damage to the endothelium. Thus, exposure of cells to random velocity fluctuations in regions of quasi-static flow compromises the integrity of the monolayer. Identification of such sites and acquisition of the knowledge necessary to protect the cells from denudation will be valuable for the endothelialization efforts of cardiac prostheses.

Successful Dynamic Cardiomyoplasty with Pharmaceutical Support

Dynamic cardiomyoplasty is an attractive alternative to heart transplantation. We used fibrin sealant to facilitate the intraoperative bonding of skeletal muscle to the myocardial wall, focusing on prevention of ischemia-reperfusion injuries in the skeletal muscle flap, and enhancement of angiogenesis in the “repaired” heart. In a sheep model, we used autologous fibrin sealant to join the tissues, to create a provisional matrix for angiogenesis, and to act as a depot for the delivery of agents aimed at minimizing ischemia-reperfusion lesion formation. Coadministered with the fibrin sealant were the following pharmaceuticals: deferoxamine (an iron chelator), pyrrolostatin (a free radical scavenger), and aprotinin (a protease inhibitor). Five days after cardiomyoplasty, the skeletal muscle was stimulated with a progressive electrical regimen. After two months, the skeletal muscle showed none of the signs of necrosis or ischemia-reperfusion damage seen in the controls. The layer of fibrin sealant rapidly (<2 weeks) became densely vascularized with capillaries. The expedited angiogenesis provided an organic bridge between skeletal muscle and myocardium. By contrast, in controls there was poor contact between the tissues, with evidence of fiber deterioration and loss of vascular network integrity in the transposed muscle flap. Even greater angiogenic stimulation was seen when pharmaceuticals were included into the fibrin meshwork, which minimized the formation of ischemia-reperfusion lesions. Over time, these agents promoted much more extensive vascularization than did fibrin sealant alone. This therapeutic strategy, using a pharmacologically-enhanced fibrin sealant, is clearly beneficial for countering muscle flap postoperative injury, and may open promising pathways for the design of other biomechanical assist devices.

Factitious Angiogenesis: Not so Factitious Anymore? The Role of Angiogenic Processes in the Endothelialization of Artificial Cardiovascular Prostheses

Due to genetic predisposition and, even more frequently, due to habitual mistreatment, (e.g., by malnutrition, smoking, etc.), the human cardiovascular system is particularly prone to massive failure, (occlusion of the blood vessels due to atherosclerosis or thromboembolism, cardiac insufficiency due to myocardial infarct, congestive heart failure), and repair, (angioplasty, cardiac transplantation). Both the anatomical repair of malfunctioning blood vessels and/or their replacement by grafting autologous blood-conduits are only partially successful: a large percentage (up to 30%) of all angioplasty as well as (coronary) bypass procedures eventually fail mainly due to re-stenosis. Similarly, although the percentage of 5-year survivors after cardiac transplantation has dramatically risen since the introduction of potent immunosuppressants, there are increasing signs of long-term adverse effects of these drugs on other organs. Furthermore, the rejection process per se is not abrogated, and often it manifests itself in the form of accelerated atherosclerosis in the blood vessels of the transplanted tissues (Mills et al. 1992). Last but certainly not least, the availability of suitable donor hearts is extremely limited, so that, eventually, most of the potential cardiac transplant patients will never receive a donor heart and die while on the waiting list. Thus, there is an increasing world-wide demand for permanent cardiovascular prostheses, such as vascular grafts, cardiac assist devices or total artificial hearts. However, currently, the long-term use of such prostheses poses serious problems, such as neointimal hyperplasia and occlusion, sepsis, thromboembolism, and calcification. Most of these complications can be traced back to the inadequate hemocompatibility of the (polymeric) blood contacting surfaces in these devices. At the previous NATO ASI on angiogenesis we delineated the ongoing attempts by us and others to restore “nature’s biocompatible blood container” through “factitious angiogenesis”, i.e., by precoating the luminal surfaces of these prostheses with a functional, nonthrombogenic monolayer of autologous endothelial cells (ECs) prior to implantation (Lelkes et al.1992). In this chapter we discuss new evidence, indicating that endothelialization of the blood contacting surface in some cardiovascular prostheses may occur through “genuine” angiogenic processes.

Durability of Endothelial Cell Monolayers Inside a Beating Cardiac Prosthesis

Thromboembolic complications associated with the use of cardiac prostheses might be alleviated by lining the blood-contacting surfaces of these devices with a functional monolayer of endothelial cells. In the current study, we tested our hypothesis that precoating textured surfaces of artificial ventricles with various plasma proteins could enhance the resistance of endothelial cell monolayers to hemodynamic forces generated within an in vitro mock circulatory loop system. Bovine jugular vein endothelial cells were grown to confluence on the luminal surface of artificial ventricles constructed of textured, medical grade polyurethane (Biospan), which had been precoated with either fibronectin or plasma cryoprecipitate. Following 7 days of culturing under static conditions, the endothelialized ventricles were connected to a mock loop system, and exposed to pulsatile flow for 6 and 24h (60bpm, 3.21/min mean flow rate, 150mmHg ejection pressure). Retention of endothelial cells was evaluated by Alamar Blue assay before and after each run. Monolayer integrity and additional morphometric parameters were also assessed by direct visualization, employing various light and electron microscopic techniques. In ventricles which had been precoated with fibronectin, Alamar Blue assay indicated cellular retention to be 77% ± 4% and 72% ± 5% of static controls, after 6 and 24h, respectively. In marked contrast, cryoprecipitate-coated ventricles retained over 90% of their endothelial cell lining through 24 h of exposure to physiological hemodynamic conditions. These findings were confirmed by visual inspection. Our study demonstrates the feasibility of maintaining an intact endothelial surface in a beating ventricular prosthesis, and that the durability of the cell layer is highly dependent upon the selection of biomaterial surface topography and protein coating.

Factitious Angiogenesis III: How to Successfully Endothelialize Artificial Cardiovascular Bioprostheses by Employing Natural Angiogenic Mechanisms

In this series of NATO ASI meetings on angiogenesis, we introduced the concept of “factitious angiogenesis” to describe a novel approach in tissue engineering aimed at the generation of permanent, hemocompatible blood conduits (31,32). We define as blood conduits a variety of cardiovascular prostheses, such as artificial vascular grafts, ventricular assist devices and total artificial hearts, and skeletal muscle ventricles. Without belaboring the profound technical and surgical problems associated with their manufacture, implantation, and/or long term use, all of these cardiovascular prostheses share a major, common obstacle: the inadequate hemocompatibility of their blood-contacting surf aces, which are made of various types of biopolymers (23,30,52). We hypothesized that the hemocompatibility in these novel blood conduits can be significantly improved by lining their blood-contacting surfaces with a non-thrombogenic monolayer of autologous endothelial cells (ECs).

Successful endothelialization of cardiovascular prostheses

Hemocompatibility of the blood-contacting surface of artificial cardiac prostheses remains a major challenge for their long-term clinical use. We hypothesized that the thrombogenicity of segmented polyurethanes (PUS) used for creating the blood sacs can be reduced by lining the luminal surface with autologous endothelial cells (EC). Over the past years we have optimized our approach by 1. choosing an appropriate (detoxified) biomaterial (Biospan@), 2. generating a controlled, roughened PU surface and 3. precoating the luminal surface with an autologous protein complex (APC) made from plasma cryoprecipitate, 4. using autologous EC isolated from adipose tissue or the jugular vein, 5. designing a seeding device with 3-D rotation, which allows for the effective, uniform endothelialization of fully assembled cardiac prostheses,and 6. developing techniques for non-destructive monitoring of the extent of EC coverage. In vitro studies confirm the importance of exposing the EC lining the blood sacs to hemodynamic forces (flow and cyclic strain) prior to implantation for establishing a nonthrombogenic, shear-resistant EC monolayer. We recently tested fully endothelialized Ventricular Assist Devices (VADs) in vitro in a mock-loop circulation (3.2 l/min flow at 1 Hz) for 6 and 24 hours. At 6 hours, no denudation was observed. After 24 hours the integrity of the EC-monolayer on fibronectin-coated PUS was partially (<300/o) impaired. By contrast, virtually no denudation was observed when the cells were seeded on Biospan@ precoated with bovine cryoprecipitate. Similar results were obtained in our first er vivo tests in which fully endothelialized VADs were connected to an extracorporeal aorta-to aorta bypass in a calf model. Our results provide the first evidence for sucecssful in vitro and w vivo testing of endothelialized VADs and stress the importance of the composition of the extracellular matrix for maintaining a durable EC monolayer.