The lymphatic system plays an important role in fluid and protein balance within the interstitial spaces. Its dysfunction could result in a number of debilitating diseases, namely lymphedema. Lymphatic vessels utilize both intrinsic and extrinsic mechanisms to pump lymph. Intrinsic pumping involves the active contraction of vessels, a phenomenon that is regulated in part by nitric oxide (NO) produced by lymphatic endothelial cells (LECs). NO production by arterial endothelial cells has been shown to be sensitive to both shear stress and stretch. Therefore, because of the unique mechanical environment of the LECs, we hypothesize that mechanical forces play an important role in regulation of the lymphatic pumping. Parallel-plate flow chambers and indenter-based cyclic stretch devices were constructed and used to apply mechanical loads to LECs. In addition, high-throughput micro-scale channels were developed and tested for shear experiments to address the need to increase the productivity and high- resolution imaging. Twenty-four hours treatment of LECs with different shear stress conditions showed a shear-dependent elevation in NO production. Moreover, 2.5 folds increase in cumulative NO was observed for stretched cells compared to the unstretched cells over six hours period. In conclusion, the upregulation observed in NO production under mechanical stimuli suggest new regulatory mechanisms that can be pharmaceutically targeted. These results provide an unprecedented insight into lymphatic pumping mechanism.

The number of people having diabetes mellitus is increasing worldwide at an alarming rate. An unbalanced diet rich in carbohydrates and saturated fats, obesity and lack of physical activity, are being blamed. The worldwide prevalence of diabetes for all age-groups has been estimated to be 2.8% in 2000 and projected to be 4.4% by the year 2030. The pathogenesis of diabetes, especially the recent epidemic increase in type 2 diabetes, is still far from clear. Endothelial dysfunction, commonly defined as reduced endothelium-dependent relaxation due to reduced availability of the vasodilator mediator nitric oxide (NO), is a hallmark of diabetes mellitus. Methylglyoxal (MG) is a highly reactive dicarbonyl compound mainly formed as an intermediate during glycolysis. MG is also formed to a lesser extent from protein and amino acid metabolism. However, the relative contribution of various metabolic precursors to MG formation is not known. Levels of MG have been found to be elevated in diabetic and hypertensive conditions but it is not known whether MG is the cause or the effect of these pathological conditions. The aim of my project was (i) to quantify the amount of MG and oxidative stress produced from various substrates in cultured A10 vascular smooth muscle cells (VSMCs), (ii) to investigate the acute in vivo effects of a single dose of MG on glucose tolerance in male Sprague-Dawley (SD) rats, (iii) to investigate the effects of MG on endothelial function and (iv) to investigate the effects, and the underlying molecular mechanisms, of chronic administration of MG on glucose homeostasis in male SD rats. The results show that aminoacetone, a protein metabolism intermediate, is the most potent substrate for MG formation on a molar basis, whereas D-glucose and fructose are equipotent. I also established optimum sample preparation protocols for reproducible measurement of MG in biological samples by high performance liquid chromatography (HPLC). In normal SD rats a single acute dose of MG induced glucose intolerance, reduced adipose tissue glucose uptake and impaired the insulin signalling pathway, which was prevented by the MG scavenger and advanced glycation end product (AGE) breaking compound, alagebrium (ALT-711). MG and high glucose (25 mM) induced endothelial dysfunction in rat aortic rings and cultured endothelial cells by reducing endothelial nitric oxide synthase (eNOS) phosphorylation at Ser-1177, activity and NO production. MG and high glucose also increased oxidative stress and further reduced NO availability in rat aortic rings and cultured endothelial cells. Chronic administration of MG in normal SD rats by continuous infusion with a subcutaneously implanted minipump for 28 days (60 mg/kg/day), induced metabolic and biochemical abnormalities of glucose homeostasis and insulin regulation that are characteristic of type II diabetes. In MG treated rats, insulin stimulated glucose uptake in adipose tissue, and glucose stimulated insulin release from freshly isolated pancreas, were significantly reduced as compared to saline treated control rats. At a molecular level, insulin gene transcription was significantly impaired and apoptosis and DNA fragmentation were more prevalent in the pancreas of MG treated rats as compared to untreated control rats. All of these in vivo effects of MG were attenuated by the MG scavenger, alagebrium. Our data strongly indicate that MG is a causative factor in the pathogenesis of endothelial dysfunction and type 2 diabetes mellitus.

The endothelium is an integral mediator of vascular homeostasis and a dysfunctional endothelium is now recognized as an early marker of atherosclerosis. Importantly, the non-invasive measurement of endothelial function by flow-mediated dilation (FMD) predicts future cardiovascular events. However, the appropriate method of its assessment and the mechanisms that govern FMD are still poorly understood. We investigated alternative parameters and methods of FMD measurement in healthy volunteers and cardiovascular disease patients. We found time to peak FMD to be highly variable both within and between individuals. Accordingly, continuous arterial diameter measurement post-cuff release was more sensitive in discriminating between health and disease compared to the measurement of diameter at 60’’ post-cuff release. Reperfusion to an ischemic tissue can paradoxically contribute to endothelial dysfunction development and further tissue damage, in a phenomenon known as ischemia and reperfusion (IR) injury. Previous exposure to sublethal ischemia (ischemic preconditioning (IPC)) can reduce sensitivity to IR injury and pharmacologic agents have since been shown to mimic this response. Using the FMD technique, we investigated various preconditioning strategies to prevent IR-induced endothelial dysfunction in the forearm vasculature of healthy volunteers. The sodium-hydrogen exchanger inhibitor amiloride and the angiotensin-converting enzyme inhibitor captopril were found not to provide endothelial protection from IR. In contrast, potent protection from IR-induced endothelial dysfunction was observed during the high-estrogen, late follicular phase of the menstrual cycle in pre-menopausal women. Finally, daily episodes of IPC were found to provide endothelial protection equipotent to an acute episode of IPC. The findings from the FMD methodological study highlight the importance of continuous arterial diameter measurement post-cuff deflation, and provide mechanistic insight that may contribute to measurement standardization and normalization. The results of the preconditioning studies improve our understanding of potential approaches to mitigate the detrimental effects of IR on the endothelium in humans.

Modular tissue engineering is a novel approach to creating scalable, self-assembling three-dimensional tissue constructs with inherent vascularisation. Under initial methods, the subcutaneous implantation of human umbilical vein endothelial cell (HUVEC)-covered collagen modules in immunocompromised mice resulted in significant host inflammation and limited HUVEC survival. Subsequently, a minimally-invasive injection technique was developed to minimize surgery-related inflammation, and cell death was attributed to extensive apoptosis within 72 hours of implantation. In confirmation of in vitro results, coating collagen modules with fibronectin (Fn) was shown in vivo to reduce short-term HUVEC apoptosis by nearly 40%, while increasing long-term HUVEC survival by 30% to 45%. Consequently, a 100% increase in the number of HUVEC-lined vessels was observed with Fn-coated modules, as compared to collagen-only modules, at 7 and 14 days post-implantation. Furthermore, vessels appeared to be perfused with host erythrocytes by day 7, and vessel maturation and stabilization was evident by day 14.

Endothelial Progenitor Cells (EPCs) are thought to participate in the healing of injured vascular endothelium by incorporating into the defect sites to mediate endothelial recovery. Recently, osteopontin (OPN) was shown to be fundamental in accelerating estrogen-dependent healing of injured blood vessels. Here, we are investigating the effect OPN has on EPC behavior. Late outgrowth human EPCs (LEPCs) were derived from circulating monocytes isolated by leukophoresis, and grown in culture until passage six. L-EPCs were then assayed for adhesion, spreading, chemotaxis, and haptotaxis, as well as resistance to detachment by flow electric cellsubstrate impedance sensing (ECIS). The results of standard and ECIS methods showed both dose and time dependent responses in cell adhesion and spreading. In addition, OPN promoted haptotactic migration of EPCs in Boyden chamber assays. LEPCs seeded onto 10μM OPN substrates and exposed to laminar flow had grater survival and higher resistance to detachment than OPN/static and flow only conditions. CD44 and !1 integrins were only responsible for approximately 50% of LEPCs adhesion to OPN compared to the unblocked condition. Western blots showed that Rho GTPases were activated in L-EPCs seeded on OPN. However, this activation could not be completely blocked by either CD44 or !1 integrin antagonists. These data confirm the direct effects of OPN on EPCs adhesion, and suggest that OPN works by mediating cell adhesion during vascular injury.

Diabetes mellitus is a chronic disease associated with vascular complications. Vascular endothelial dysfunction caused by increased endothelial cell apoptosis contributes to diabetic cardiovascular complications. The glucose metabolic by-product, D-lactate, is elevated in diabetics and it is unknown whether it contributes to endothelial cell apoptosis. We hypothesized that diabetic D-lactate levels induce apoptosis in human vascular endothelial cells (HUV-EC-C). HUV-EC-C were incubated with 0.2 mM D-lactate (DLA) and mRNA expression of PI3K/AKT pathway members (AKT1, Bcl-2, BAD, eNOS, PI3K) were measured using Quantitative RT-PCR. DLA downregulated all genes at 6 and 24 hours, followed by increase in expression after 48 hours except PI3K, which remained below control. To further investigate apoptosis, the Human Apoptosis PCR Array was used and expression of all proapoptotic genes (TNF family members) and antiapoptotic genes (IAP family members) were decreased and increased, respectively, at 24 hours followed by an increase and decrease, respectively, at 48 hours. Caspase activity, measured using the Caspase-Glo® 3/7 Assay after HUV-EC-C exposure to 0.2 mM DLA alone or in combination with 20 mM glucose (GLU) or 5 µM methylglyoxal (MG), was increased after 1, 72, and 96 hours. Furthermore, to know whether DLA (0.2 mM) and DLA (0.2 mM), GLU (20 mM) and MG (5 µM) combined cause changes in cellular energy metabolism, creatine (Cr) and high-energy phosphate substrates (CrP, ATP, ADP, AMP) were quantified using HPLC and no changes were observed. We further measured ROS production in HUV-EC-C treated with 0.06-2 mM DLA alone or 0.2 mM DLA with 5-30 mM GLU or 5-160 μM MG. All DLA concentrations increased ROS production by 160% to 216%. DLA with GLU or MG significantly increased ROS production compared to GLU or MG alone. Lastly, D-lactate dehydrogenase (D-LDH) expression was determined using Quantitative RT-PCR and D-LDH was not detected in HUV-EC-C. In conclusion, DLA altered expression of different pro- and anti-apoptotic genes in HUV-EC-C. Furthermore, exposure of HUV-EC-C to DLA levels typically present in diabetics resulted in time-dependent changes in caspase activity, possibly due to excessive ROS production. Whether these changes eventually lead to endothelial dysfunction in diabetes needs further investigation.

Transcapillary exchange is the movement of fluid and molecules through the porous capillary wall, and is important in maintaining homeostasis of bodily tissues. The classical view of this process is that of Starling's hypothesis, in which the forces driving filtration or absorption are the hydrostatic and osmotic pressure differences across the capillary wall. However, experimental evidence has emerged suggesting the importance of the capillary wall ultrastructure, and thus rather than the global differences between capillary and tissue, it is the local difference across a structure lining the capillary wall known as the endothelial glycocalyx that determines filtration. Hu and Weinbaum presented a detailed cellular level microstructural model of this phenomenon which was able to explain some experimental discrepancies. In this Thesis, rather than describing the microstructural details, the capillary wall is treated as a poroelastic material. The assumptions of poroelasticity theory are such that the detailed pore structure is smeared out and replaced by an idealized homogeneous system in which the fluid and solid phases coexist at each point. The advantage of this approach is that the mathematical problem is greatly simplified such that analytical solutions of the governing equations may be obtained. This approach also allows calculation of the stress and strain distribution in the tissue. We depart from classical poroelasticity, however, due to the fact that since there are concentration gradients within the capillary wall, the filtration is driven by both hydrostatic and osmotic pressure gradients. The model predictions for the filtration flux as a function of capillary pressure compares favourably with both experimental observations and the predictions of the microstructural models. An important factor implicated in transcapillary exchange is the endothelial glycocalyx, which was shown experimentally to protect against edema formation. Using our theory in combination with the experimental measurements of glycocalyx thickness and pericapillary space dimension (PSD), we make a quantitative comparison for the excess flow as a result of a deteriorated glycocalyx, which shows reasonably good agreement with the data. Since many of the parameters in the model are difficult to measure, a sensitivity analysis was performed on the most important of these. Finally, since there was variation in the measurements of glycocalyx thickness and PSD, we used probability distributions to represent the data, and performed further calculations to obtain ranges of likely values for the various parameters. This work could find applications in cardiovascular disease, where the glycocalyx is degraded or absent, and in cancer research, where the abnormal vasculature is an impediment to the efficient delivery of anti-cancer drugs.

<p>The fatty acid linoleic acid has been identified as a potential mediator of atherosclerotic plaque development. Treatment of monocytes with linoleic acid leads to an increase in monocyte adhesion to endothelial cells under flow conditions; however, the mechanisms through which linoleic acid affect monocyte adhesion remain unclear. Using a combination of micropipette aspiration techniques and fluorescent microscopy, I tested the hypothesis that linoleic acid increases membrane tether formation between monocytes and endothelial cells. </p><p>Treatment of U937 monocytes with free linoleic acid or albumin-bound linoleic acid reduced the cortical tension of the monocytes. The effects of albumin-bound linoleic acid on the membrane were governed by the exchange of linoleic acid from albumin to the membrane and by the removal of fatty acids from the membrane by fatty acid binding sites on albumin. </p><p>The frequency of tether formation between U937 monocytes and TNF-α stimulated HUVECs increased following treatment with free linoleic acid or albumin-bound linoleic acid. The increase in tether frequency was not due to an increase in monocyte deformability or adhesion receptor expression. Tether extraction occurred primarily through E-selectin. Treatment with free linoleic acid increased the localization of E-selectin to clathrin-coated pits suggesting an increase in the formation of nanoclusters of E-selectin on HUVECs. The increase in tether frequency was blocked by the U73122 phospholipase C inhibitor indicating that linoleic acid increased monocyte adhesion through a phospholipase C mediated mechanism.</p><p>Treatment with free linoleic acid did not affect the threshold force for tether extraction or the effective viscosity of tethers extracted from HUVECs, but it decreased the threshold force for tether extraction from U937 monocytes and increased the effective tether viscosity. Treatment with U73122 blocked the reduction in the threshold force indicating that linoleic acid affected the regulation of the membrane adhesion energy through the hydrolysis of PIP2 by phospholipase C.</p><p>The results of the study indicated that linoleic acid promoted membrane tether formation by increasing E-selectin bond formation and reducing the adhesion energy between the U937 plasma membrane and the actin cytoskeleton through the hydrolysis of PIP2 by phospholipase C.</p> / Dissertation

Aortic valve disease (AVD) affects millions of people of all ages around the world. Current treatment for AVD consists of valvular replacement with a non-living prosthetic valve, which is incapable of growth, self-repair, or remodeling. While tissue engineering has great promise to develop a living heart valve alternative, success in animal models has been limited. This may be attributed to the fact that understanding of valvular cell biology has not kept pace with advances in biomaterial development. Aortic valve leaflets are exposed to a complex and dynamic mechanical environment unlike any in the vasculature, and it is likely that native endothelial and interstitial cells respond to mechanical forces differently from other vascular cells. The objective of this thesis was to compare valvular cell phenotype to vascular cell phenotype, and assess the influence of steady shear stress on valvular cell biology. This thesis demonstrates that valvular endothelial cells respond differently to shear than vascular endothelial cells, by aligning perpendicular to the direction of steady shear stress, and by the differential regulation of hundreds of genes in both static and fluid flow environments. Valvular interstitial cells expressed a combination of contractile and synthetic phenotypes not mimicked by vascular smooth muscle cells. Two three-dimensional leaflet models were developed to assess cellular interactions and the influences of steady laminar shear stress. Valvular co-culture models exhibited a physiological response profile, while interstitial cell-only constructs behaved more pathologically. Steady shear stress enhanced physiological functions of valvular co-cultures, but increased pathological response of interstitial cell-only constructs. These results showed that valvular cells, whether cultured separately or together, behaved distinctly different from vascular cells. It was also determined that shear stress alone cannot induce tissue remodeling to more resemble native valve leaflets. The leaflet models developed in this thesis can be used in future experiments to explore valvular cell biology, assess the progression of certain forms AVD, and develop targeted diagnostic and therapeutic strategies to hopefully eliminate the need for valvular replacement entirely.

Sickle cell anemia is a hematological disorder that is caused by a single point mutation in the beta-globin chain of hemoglobin. It results in several complications related to the small and large vessels in patients with the disease. Large vessel complications include cerebral infarcts, which are observed in children under ten years old. The mechanism behind this complication is not completely understood. It is the goal of this project to begin to understand the role sickle erythrocytes may play in causing endothelial dysfunction as a precursor to sickle related complications. The hypothesis of this work is that exposure of large vessel endothelium to sickle erythrocytes causes an increase in endothelial permeability through loosening of adherens junctions. In the first goal of this work, bovine aortic endothelial cells (BAECs) are grown on coverslips and exposed to sickle erythrocytes for 5 minutes and either immediately fixed or incubated in 30 minutes and then fixed. Immunofluorescent studies labeling VE cadherin show changes in VE cadherin dynamics, suggesting sickle erythrocytes may be involved in this observation. Next, BAECs were grown on transwell inserts and exposed to sickle erythrocytes for 5 minutes. The erythrocytes are washed off and the BAEC are incubated with 10,000 MW dextran conjugated to lucifer yellow or FITC-BSA or to determine BAEC permeability. When dextran is used as the test molecule, endothelial permeability did not show a significant change from baseline. However, when BSA is used as the test molecule, increases in endothelial permeability are observed. Explanations into the differences between the transport mechanisms of the two molecules are discussed. These experiments show changes in VE cadherin localization due to sickle erythrocyte exposure. This may cause increases in endothelial permeability and an experimental model and preliminary studies are performed. This study provides potential mechanisms to explain the changes in VE cadherin localization and provide suggestions for further studies to test the effect of sickle erythrocytes on endothelial permeability. This work provides a strong foundation for continuing studies on the effects of sickle erythrocytes on endothelial dysfunction within the confines of sickle related complications.