Vascular endothelium – Gatekeeper of vessel health

Abstract

The vascular endothelium is an interface between the blood stream and the vessel wall. Changes in this single cell layer of the artery wall are believed of primary importance in the pathogenesis of vascular disease/atherosclerosis. The endothelium responds to humoral, neural and especially hemodynamic stimuli and regulates platelet function, inflammatory responses, vascular smooth muscle cell growth and migration, in addition to modulating vascular tone by synthesizing and releasing vasoactive substances. Compromised endothelial function contributes to the pathogenesis of cardiovascular disease; endothelial ‘dysfunction’ is associated with risk factors, correlates with disease progression, and predicts cardiovascular events. Therapies for atherosclerosis have been developed, therefore, that are directed towards improving endothelial function.

1. Atherosclerosis

Atherosclerosis is a chronic inflammatory disease in which the artery wall thickens as a result of the accumulation of cholesterol, macrophages and smooth muscle cells (SMC), ultimately restricting blood flow through the artery. It is the main pathologic condition underlying coronary artery and cerebrovascular disease leading to heart attack and stroke, respectively. In the pathogenesis of atherosclerosis (reviewed in Ref. [1–4]), increases in plasma low density lipoprotein (LDL) leads to a proportional increase in the entry of cholesterol laden LDL particles into the arterial wall across a ‘compromised/dysfunctional’ endothelial monolayer, where it accumulates. Once there, it can become oxidized, by free radical production from adjacent endothelium, smooth muscle cells or isolated macrophages [5–8]. Oxidized LDL has numerous effects on a variety of cells, many of which are believed to cumulatively exacerbate atherothrombosis ([9] for review). These include promotion of monocyte adhesion and infiltration to the intima by causing production of monocyte chemotactic protein-1 (MCP-1) by endothelium and expression by endothelium of monocyte-binding proteins including intercellular adhesion molecule-1 (ICAM-1), foam cell formation following uptake of oxidized LDL via scavenger receptors (SR-A type I and II and CD36), and stimulation of the migration of medial SMC into the intima where they proliferate in response to growth factors such as platelet derived growth factor (PDGF) [13]. In the intima, SMC produce extracellular matrix molecules including collagen and elastin. The most common clinical complication of atherosclerosis occurs upon plaque rupture that allows blood components to come into contact with plaque lipids and tissue factor, resulting in thrombus formation. While several cell types are clearly involved in the pathogenesis of atherosclerotic plaques, endothelial compromise/dysfunction is deemed of particular importance as it is a necessary and initiating occurrence for atherogenesis to proceed. The endothelial lining can be viewed as the first line of defense between risk factors and vascular disease.

2. Arterial endothelium: structure, function

The vascular endothelium lines the entire circulatory system. In coronary arteries, this single layer, together with some extracellular matrix, comprises the tunica intima. Originally thought of simply as a passive barrier, it is now viewed as an organ whose normal functioning is crucial to maintaining vascular health, and whose dysfunction is key in the initiation, progression and clinical complications of vascular disease. Vascular endothelium acts as a selectively permeable barrier between extravascular and intravascular compartments and provides a nonthrombogenic lining for the cardiovascular system. It is poised in an anatomic location that allows interaction not only with circulating blood components and cells, but also with cells in the vessel wall. Endothelial cells are long, flat cells orientated in the long axis of the vessel. An average endothelial cell is 20–40 μM long, 10–15 μM wide and only 0.1–0.5 μM thick. Electron microscopy of arterial endothelium reveals it to be of the continuous type characterized by tight junctions at the lateral borders of each cell that restrict the movement of macromolecules, and a complex micro-vesicular system implicated in macromolecular transport. Despite its apparent morphological simplicity and relative homogeneity, there is evidence of regional and species variation manifested by differences in permeability, responsiveness and biosynthesis.

3. Barrier function

The role of the endothelium as a semipermeable barrier is one of its most basic functions. It regulates transport of macromolecules between the vascular lumen and vascular smooth muscle. There are several different mechanisms by which macromolecules cross the endothelial barrier; through the endothelial cells themselves, by diffusing laterally within the endothelial cell membrane and thus around the cell, through endothelial cell-to cell junctions, through endothelial gaps, or via vesicular transport. As most biologic molecules are large, hydrophilic, anionic molecules, which are unable to diffuse into and through bilayer membranes, most are thought to move through intercellular junctions between cells or by vesicular transport, or the formation of transient channels resulting from vesicle fusion [10]. Reorganization of the inter-cellular junctions, involving actin and myosin or direct dissolution of junctional contacts, is believed to be the primary mechanism by which endothelial permeability to water, small and large solutes is increased [11].

4. Changes in permeability

Various physiologic and pathophysiologic stimuli can induce, acutely and chronically, dramatic changes in endothelial permeability. For example thrombin, histamine and other acute inflammatory mediators can act on endothelium to stimulate opening of their intercellular junctions at the level of adherens and tight junctional complexes [12]. The signaling involved in mediating these responses include protein kinase C (PKC)–induced phosphorylation of linking proteins at the cell-cell and cell-matrix junctions, leading to actin reorganization, cell rounding and increased paracellular transport. Myosin light chain kinase (MLCK) may also be activated by inflammatory mediators leading to actin-myosin based retraction of endothelial cells [12]. Vascular endothelial growth factors (VEGFs) are key regulators of vascular permeability via nitric oxide synthase (NOS) regulation [13,14], facilitated by multiple mechanisms including gap formation and vesiculo-vacuolar organelle formation [15] (reviewed in Ref. [16]). The bacterial endotoxin lipopolysaccharide (LPS) causes endothelial hyperpermeability by stimulating the small GTPase, RhoA, and its effector Rho Kinase (ROCK) [17]. Endothelial permeability is also influenced by fluid shear stress, the tangential frictional force exerted by flowing blood [10,18]. Investigation of the relationship between albumin permeability and shear stress magnitude in vivo revealed that endothelium exposed to low wall shear stresses was more likely to have elevated macromolecule permeability [19]. Of importance with respect to vascular disease, low density lipoprotein (LDL) accumulation is greater at these more permeable areas exposed to low wall shear stress [20,21]. It is now recognized that atherosclerotic lesions preferentially develop at vessel segments exposed to low, disturbed or oscillating flow, while laminar flow and high shear stress are seemingly atheroprotective [22]. In addition to inhibition of thrombosis and inhibition of endothelial apoptosis, limitation of permeability is now a well-characterized atheroprotective mechanism of laminar flow and high shear stress [22]. Conversely, at regions of pathological low shear stress, increased endothelial permeability results in enhanced infiltration of LDL and its local accumulation, which is a critical initial event in the development of atherosclerosis [23,24]. Rozenberg et al., recently reported that Histamine, acting via its H1 receptor drives the formation of atherosclerotic lesions through an increased vascular permeability for LDL [25]. Mullick et al., showed that exposure to cigarette smoke injures the endothelium, resulting in increased arterial permeability and increased LDL accumulation [26]. A recent study provided evidence of a role for lipoxygenase and its metabolite hydroxyeicosatetraenoic acid (HETE) in ‘high fat diet’-induced endothelial tight junction disruption [27], thus providing a possible mechanistic link between lifestyle and atherosclerosis. Therefore, a wide variety of injurious stimuli (e.g., certain hemodynamic forces, inflammatory mediators, bacterial endotoxin LPS, environmental toxins, high fat diet) can contribute to endothelial dysfunction by increasing endothelial permeability and subsequently arterial lipid accumulation in the subendothelial space, thereby initiating atherosclerotic plaque development. On the other hand, factors shown to maintain endothelial barrier function include high density lipoprotein (HDL), and physical exercise (reviewed in [28]). With respect to the latter, there is considerable evidence of a direct relationship between exercise and vascular health. The effect of exercise on maintaining endothelial barrier function is likely a consequence of exercise increasing blood flow and shear stress, which in turn releases ‘vasoprotective’ molecules such as nitric oxide (NO) and prostacyclin (PGI₂). It is also appreciated that Sphingosine-1-phosphate (S1P), a bioactive sphingolipid associated with HDL and found mainly in the blood and lymph, robustly promotes endothelial barrier function [29,30]. In particular, growing evidence indicates that HDL-associated S1P mediates the beneficial effects on endothelial integrity [31,32]. Other endogenous factors known to increase endothelial barrier function and decrease permeability are angiopoietin-1 [33], and the second messenger cyclic adenosine monophosphate (cAMP) and agonists such as Serotonin and β-adrenergic agonists that increase it [34,35] (Table 1).

Table 1.

Factors affecting endothelial cell permeability.

Increase permeability	Decrease permeability
'Low' shear stress	'High' shear stress
Vascular endothelial growth factor (VEGF)	High density lipoprotein (HDL)/Sphingosine 1 phosphate(S1P)
Lipopolysaccharide (LPS)
High fat diet	Physical exercise
Phospholipid oxidative products	Increased cellular cAMP (by e.g., Serotonin or β-adrenergics)
Inflammatory mediators (e.g., thrombin, Histamine, TGF(β) Cigarette smoke	Angiopoietin-1

5. Endothelial glycocalyx

Consisting of a negatively charged, organized mesh of membranous glycoproteins, proteoglycans and associated plasma proteins, the endothelial glycocalyx (100–750 nm thick) is recognized as contributing, together with the endothelium, to the protection of the vascular wall against disease (for review [36,37]). Improved fixing and imaging techniques have allowed in vivo visualization of the endothelial glycocalyx and provided evidence that there are significant reductions in its extent during acute and chronic inflammatory challenge in man. Indeed, destruction of the glycocalyx has been directly and indirectly evidenced in several studies, e.g. after ischemic challenge [38–40], during redox stress [39], and after inflammation [40,41]. Deterioration or destruction of the glycocalyx is associated with pathophysiological sequelae including increased endothelial permeability, platelet aggregation and loss of vascular responsiveness [42]. There are interesting data supporting an important role for the glycocalyx in the initiation and progression of atherosclerosis. Van den Berg et al., reported thinning of the glycocalyx in mice fed a cholesterol rich diet, and demonstrated less glycocalyx in regions of the vessel at high atherogenic risk (i.e., at branch points, areas of curvature) [43]. Moreover, the authors found an inverse relation between glycocalyx thickness and the intima-media ratio [43]. These findings support the hypothesis that diminution of the glycocalyx, perhaps as a consequence of turbulent flow/low shear stress, contributes to the vulnerability of regions that are at high atherogenic risk [44]. Vasculoprotective effects attributed to the glycocalyx include fortifying endothelial barrier function [45], inhibiting coagulation and leukocyte adhesion [46] and mediating shear stress-induced NO release [47,48]. The seemingly atheroprotective role of the gylcocalyx means that it is attracting attention as a potential novel diagnostic or therapeutic target for atherosclerosis [49,42]. Approaches to increase the glycocalyx or to reverse its damage, as well as tactics to maintain it mechanically and chemically, are being tested as viable clinical options at this stage [37,42]. Indeed, clinical trial data support a protective role for the compounds Diazoxide and Sulodexide on endothelial glycocalyx in patients after coronary bypass grafting and in those with type 2 diabetes, respectively [50,51].

6. Hemodynamic forces

Cells of the arteries are continuously exposed to pulsatile blood flow and are therefore subjected to the hemodynamic forces shear stress and cyclic strain. Shear stress is the tangential frictional force acting predominately at the endothelial cell surface, whereas cyclic strain/stretch acts perpendicular to the vessel wall and affects both endothelial and smooth muscle cells together with the underlying matrix [52]. Shear stress (ϊ) (expressed as ‘dynes/cm²’) can be calculated using the formula, ϊ = 4μQ/πr³, where μ is blood viscosity, Q is blood flow volume, π is the ratio of the circumference of a circle to its diameter, and r is the radius of the blood vessel. The pumping heart causes pulsatile changes in blood pressure that stretch the vessel wall circumferentially and create cyclic strain in vascular cells. The degree of stretch is around 9%–12% in the aorta and 1%–2% in the carotid arteries [53]. A variety of devices have been employed by research laboratories to expose cultured cells to defined mechanical forces. Those used to apply shear stress include the cone plate viscometer [54], the parallel flow chamber [55] and the perfused capillary culture system [56,57]. In the case of cyclic stretch, researchers have grown cells on a variety of elastic substrates [58] that can then be periodically stretched; e.g., the Flexercell system [59]. In order to respond to mechanical forces such as shear stress or cyclic strain and convert it into a biochemical response endothelial cells must express special receptors or ‘mechanotransducers’. Biomechanical studies have suggested many different mechanotransducer candidates including ion channels, tyrosine kinase receptors, G-protein coupled receptors, a caveolaemediated process, cell adhesion molecules, the glycocalyx and primary cilia (Reviewed in Ref. [53,60]). Certain biomechanically induced events appear to involve transcriptional regulation and shear stress-response elements have been identified in the promoters of multiple endothelial cell-expressed genes (e.g., [61]).

Recent studies have uncovered other contenders as mechanotransducers. Syndecan 4, a transmembrane heparan sulfate proteoglycan, is required for proper flow alignment in cultured endothelial cells [62]. Hypercholesterolemic mice deficient for syndecan 4 have increased atherosclerotic plaque when compared with control mice [62]. A flow-dependent GTP exchange factor called TIAM1 reportedly links platelet endothelial cell adhesion molecule 1(PECAM-1) mechanotransduction to focal activation of the small GTPase Rac1 which in turn triggers the Nuclear factor kappa B (NF-kB) pathway and production of reactive oxygen species [63]. Another mechanotransducer candidate is the ion channel protein Piezo1, a mediator of shear stress–induced calcium influx [64] Piezo1-mediated calcium influx is important for calpain activity and subsequent rearrangement of focal adhesions for flow alignment. Endothelial cells isolated from Piezol^−/− mice failed to align properly when exposed to atheroprotective flow conditions [65] A role for the G protein–coupled Sphoingosine-1-phosphate receptor (S1P1) in endothelial mechanotransduction has also been put forward as the expression of S1P1 was critical for flow-mediated directional alignment in cultured endothelial cells in vitro and in vivo [66].

Hemodynamic forces are ubiquitous and fundamental physiologic stimuli for vascular cells, and are believed to critically influence atherogenesis by regulating endothelial cell function, smooth muscle behavior, and the interaction of endothelial cells with smooth muscle cells [67] and with leucocytes and other blood constituents [22,68,69]. Of note, flow patterns and hemodynamic forces are not uniform throughout the vasculature. While blood flow is essentially laminar throughout the arterial system, in straight parts of the arterial tree blood flow is generally steady and unidirectional and wall shear stress is high (>15 dyn/cm²; ‘physiologic flow’). At branch points and curvatures, blood flow is disturbed or oscillatory with low net wall shear stress (0–4 dyn/ cm²; ‘pathologic low flow’). Atherosclerotic lesion development generally correlates with the latter [70]. This type of observation gave rise to the concept, now accepted, that disturbed flow patterns are atherogenic whereas steady high flow patterns are atheroprotective (reviewed in Ref. [71]). Several groups have investigated the effects of various flow patterns on endothelial cell biology in vivo and in vitro over the last 30 years, and the emerging consensus is that physiologic flow favors the expression of endothelial genes and bioactive products that are protective against atherosclerosis, whereas pathologic flow stimulates genes and products that promote atherogenesis [22,68,69,70,72] (Fig. 1). Indeed, studies using DNA microarrays have revealed striking differences between the numbers and kinds of endothelial genes that respond to steady flow versus disturbed flow [73]. Of interest, it is emerging that flow modulates epigenetic DNA methylation patterns via alterations in DNA methyltransferase activity, in particular DNMT1, that may contribute to the endothelial phenotypes observed in regions of disturbed flow [74]. For example, disturbed flow patterns increase methylation of the proximal promoter of Kruppel like factor 4 (KLF4), thus inhibiting ‘atheroprotective’ KLF4 transcription in atherosclerosis-susceptible regions [75]. Findings such as these suggest that DNA methylation may play a key role in maintaining endothelial cell homeostasis and in vascular disease development and encourage continued investigation into epigenetic mechanisms relevant to vascular health and disease.

Fig. 1.

Effect of different flow patterns on (a) atherosclerotic plaque development and (b) endothelial phenotype and activity. In straight parts of the arterial tree blood flow is steady and unidirectional and wall shear stress high. At branch points and curvatures, blood flow is disturbed or oscillating and net wall shear stress is low (areas in blue in schematic of arterial bifurcation). Steady, high shear flow is atheroprotective, whereas disturbed flow, low shear is atherogenic. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

7. Hemodynamic forces and endothelium

Hemodynamic forces impact multiple aspects of endothelial cell biology and function (reviewed in Ref. [53]), all of which may influence atherosclerotic plaque initiation and development. Shear stress affects endothelial cell morphology. Endothelial cells exposed to steady rapid flow are elongated in shape and aligned with their long axis in the direction of flow, in contrast to the rounder shape and non-uniform orientation and greater permeability of EC in segments exposed to disturbed flow [76]. Shear stress plays a role in controlling endothelial cell proliferation and apoptosis; e.g., steady flow reduced EC proliferation [77], whereas disturbed flow increased EC turnover and stimulated apoptosis [78]. Cyclic strain also promotes endothelial migration and in vitro tube formation (an index of angiogenesis) [79]. Growth and migration of endothelial cells becomes important in the repair of areas denuded of endothelium as a result of injury, be it toxic or mechanical and/or iatrogenic.

Endothelial cells respond to hemodynamic forces by altering their production of vasoactive substances (Fig. 2). Increases in shear stress generally causes vasodilation, mediated for the most part by increased endothelila nitric oxide synthase (eNOS) activity and NO production [57,80]. Indeed, shear stress is considered the premiere physiologic stimulus for this potent vasorelaxant molecule. Other endothelial-derived vasoactive substances altered by shear stress include PGI₂ [81,57,82] and endothelin-1 (ET-1) [83,84]. The vasodilators NO and PGI₂, and the vasoconstrictor ET-1 are not only important in regulating vascular tone at the level of vascular smooth muscle, but in addition, these agents inhibit (NO, PGI₂), and stimulate (ET-1), SMC growth and migration [85–87]. Nitric oxide and PGI₂ are also potent inhibitors of platelet aggregation [88,89]. Besides stimulating NO and PGI₂, shear stress contributes to maintaining the endothelium as non-thrombogenic by stimulating thrombomodulin [90], heparin sulfate proteoglycans [91] and tissue type plasminogen activator (tPA) [92].

Fig. 2.

Shear stress and cyclic strain are important stimuli for endothelial cells (EC). These hemodynamic forces are sensed by mechanoreceptors that include ion channels, G protein coupled receptors, integrins, cytoskeletal components and receptor tyrosine kinases. Hemodynamic forces modulate endothelial release and expression of a myriad of substances including vasoactive agents, growth factors, cytokines, microRNAs (miRNAs), thrombomodulin and reactive oxygen species (ROS) that may affect the aggregation and adhesion of cells circulating in the blood (i.e., platelets and monocytes), the permeability, adhesivity and angiogenic potential of EC themselves, as well as the growth, migration, vasoreactivity and extracellular matrix (ECM) synthesis of the underlying vascular smooth muscle (SMC).

Hemodynamic forces also affect the production by the endothelium of a variety of growth factors and cytokines, many of which can affect the underlying smooth muscle cells, stimulating them to de-differentiate from a quiescent, contractile phenotype to the migratory and proliferative synthetic phenotype fundamental to the pathogenesis of vascular disease [93]. These include platelet-derived growth factor (PDGF) [94], basic fibroblast growth factor (bFGF) [95], transforming growth factor beta (TGFβ) [96] and Interleukin-1 and -6 [97].

Modulatory influences of shear stress on endothelial cell adhesion molecule (e.g., vascular cell adhesion molecule-1, VCAM-1) and selectin (e.g., P-selectin) expression have also been reported, effects with important implications to not only platelet aggregation [98], but also to the adhesion of leukocytes, in particular monocytes, to the endothelium [99,100]. Of note, following their adhesion, migration of monocytes across the endothelial barrier is considered a significant early event in the development of atherosclerotic lesions [1].

7.1. Reactive oxygen species

Increased production of reactive oxygen species (ROS) contributes to mechanisms of vascular/endothelial dysfunction and atherosclerosis [101,102]. ROS is a term that includes oxygen radicals such as superoxide (${\mathrm{O}}_2^{-}$) and hydroxyl radical (OH), as well as non-radical derivatives of O₂ including hydrogen peroxide (H₂O₂) and Ozone (O₃). Oxidative stress is mainly caused by an imbalance between the activity of endogenous pro-oxidative enzymes (such as nicotinamide adenine dinucleotide phosphate-oxidase (NADPH) oxidase and xanthine oxidase) and anti-oxidative enzymes (such as superoxide dismutase, glutathione peroxidase and heme oxygenase) in favor of the former. Modulation of ROS production from endothelium by shear stress and cyclic stretch has been reported [60,103]. For example, Chatterjee et al., showed that in response to a decrease of shear, such as occurs during ischemia, membrane depolarization via ATP-sensitive K+ channel closure initiated a signaling cascade leading to NADPH oxidase activation and ROS production [104]. In contrast, shear stress upregulated the expression of antioxidant enzymes such as peroxiredoxin 1 (PRX 1) [105]. The accepted idea, from evidence garnered over the last 20 years, is that flow patterns crucially regulate the ROS/NO balance, with steady flow (high shear) causing lower levels of ROS and greater NO bioavailability (considered anti-atherogenic), whereas disturbed flow (low shear) results in higher ROS and lower NO bioavailability (considered pro-atherogenic) (reviewed in Ref. [60]). Steady flow favors the activation of key transcription factors such as Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) and Krüppel-like Factor 2 (KLF2) to promote an anti-atherogenic environment by enhancing the expression of superoxide dismutase and heme oxygenase [106], whereas irregular flow leads to activation of key transcription factors such as AP-1 and NF-κB leading to enhanced expression of pro-atherogenic agents such as monocyte chemotactic protein1 (MCP-1) and intracellular adhesion molecule 1 (ICAM-1) [60].

8. Cyclic strain

As mentioned earlier, endothelial and smooth muscle cells in vivo are also exposed to cyclic strain, a circumferential stretching synchronous with cardiac cycle-induced changes in blood pressure. Like shear stress, cyclic strain is a force of physiological and pathological interest, sensed by mechanosensors, that modulates endothelial cell intracellular signaling resulting in changes in proliferation, apoptosis [107], migration [79], angiogenic potential [79,59] and matrix remodeling [108,109,103]. Cyclic strain increases under certain pathological conditions such as hypertension and congenital heart disease. Kou et al., reported that lower (5%) cyclic strain increases, but higher (20%) cyclic strain decreases, endothelial cell survival and angiogenesis via the NAD(P)H subunit p22phox pathway [110]. Physiologic cyclic strain regulated, in a p38-and receptor tyrosine kinase-dependent manner, the release of thrombomodulin a membrane glycoprotein expressed on the lumenal surface of endothelial cells that promotes anti-coagulant and anti-inflammatory properties [111]. Liu et al., reported that cyclic strain stimulates Heme oxygenase 1 (HO-1) in a manner dependent on the production of ROS and mediated by Nrf2 [107]. They also showed that cyclic strain inhibits cytokine-mediated endothelial cell apoptosis and that HO-1 underlies this cytoprotective effect [107]. Thus, there is ample evidence that hemodynamic forces, in particular shear stress and cyclic strain, are critically important stimuli for the endothelium, affecting endothelial cell function as well as modulating the endothelial release and expression of growth factors, cytokines, vasodilators, vasoconstrictors, reactive oxygen species and cell adhesion molecules - all of which impact the underlying vascular smooth muscle and/or the interaction of circulating inflammatory and immune cells with the endothelium and thereby affect atherogenesis (Fig. 2). There is also evidence that atherosclerosis risk factors themselves (e.g. hypertension, dyslipidemia, smoking) result in altered local hemodynamic/mechanical forces that could be an underlying mechanism for the progression of atherosclerosis [112].

9. Physical exercise

Of interest, physical exercise seems to be naturally antiatherogenic, preserving vascular function with aging (reviewed in Refs. [113,114]). Some data suggest that exercise inhibits plaque development and induces the regression of coronary stenosis [115,116]. Furthermore, experimental studies have revealed that exercise prevents the conversion of plaques into a vulnerable phenotype, thus averting the appearance of fatal lesions [117]. While the molecular mechanisms involved are not yet fully understood, strong evidence supports the importance of increased shear stress, as a result of exercise-induced increased heart rate and blood flow, as the primary signal acting on the endothelium produced by exercise [118], resulting in down-regulated endothelial angiotensin II type 1 receptor expression, leading to decreases in NADPH oxidase activity and superoxide anion production, which in turn decreases ROS generation, and preserves endothelial NO bioavailability and its protective anti-atherogenic effects [119]. Of further note, atherosclerosis is a chronic inflammatory disease [120] and several studies support an inverse correlation between physical exercise and serum markers of inflammation such as Creactive protein (CRP), tumor necrosis factor-α (TNFα) and Interleukin-6 [121–123,124. In addition, recent meta analysis of 28 individual studies showed that physical exercise has several non-acute effects expected to reduce atherogenesis; i.e., decreasing adhesion molecule VCAM-1 and ICAM-1 expression, reducing the angiogenic factor VEGF, and increasing endothelial progenitor cells (EPC) [125].

10. Endothelium's vasoactive role (vasodilators)

10.1. EDRF/NO

The smooth muscle response to sympathetic/parasympathetic nerve stimulation is a major contributor to maintenance of cardiovascular tone throughout the vascular tree. However, starting with the discovery in 1980 by Furchgott and Zawadski of endothelial-derived relaxing factor (EDRF) [126], identified chemically as endogenous nitric oxide (NO) [127], endothelial cells were subsequently recognized as playing a significant role in the local regulation of vascular tone. This local regulation has implications for atherogenesis. Nitric oxide is synthesized by the heme-containing, calcium and calmodulin-dependent enzyme nitric oxide synthase in endothelial cells (eNOS) from l-arginine in a reaction that produces stoichiometric amounts of l-citrulline [128,129]. Activation of NOS and release of the free radical NO, which easily diffuses across biological membranes, results in stimulation of a soluble guanylyl cyclase leading to a profound increase in intracellular cGMP within the adjacent smooth muscle cells, thus resulting in their relaxation and consequent vasodilation [129]. NO has a wide range of actions important in maintaining vascular homeostasis. In addition to causing vasodilation, it has anti-proliferative, antioxidant and anti-inflammatory properties that inhibit atherogenesis ([130] for recent review). As mentioned previously, shear stress resulting from blood flow upregulates eNOS in endothelium. Shear stress-responsive elements have been identified in the eNOS promotor sequence on cloning of the eNOS cDNA and identification of the promotor region [131]. eNOS activity can also be stimulated by receptor mediated agonists-e.g., bradykinin, acetylcholine and thrombin via palmitoylation, binding to calmodulin/calcium, displacement of caveolin and release from the plasma membrane [132].

The beneficial effect of physical exercise on cardiovascular disease is believed partly due to increased NO bioavailability [119,133]. Conversely, common risk factors for cardiovascular disease such as hypercholesterolemia, hypertension, smoking and diabetes mellitus are associated with reduced NO in the arterial wall [134] (Fig. 3). Indeed, the term ‘endothelial dysfunction’ has become synonymous with reduced biological activity of NO. Because of these findings, numerous therapies have been investigated based on enhancing NO release, thereby reversing endothelial dysfunction and preventing atherogenesis. NO donors (e.g., nitroglycerin) are frequently used in cardiovascular disease patients. NO bioavailability can be restored by moderate physical exercise, antioxidants and l-arginine, the natural precursor of NO [134]. Several small-scale studies have demonstrated that intravenous l-arginine augments endothelial function and improves exercise ability in patients with cardiovascular disease by enhancing vasodilation and reducing monocyte adhesion [135]. Other agents under investigation include tetrahydrobiopterin (BH4, a co-factor required for activation of NOS) and folic acid (which may mimic BH4). Of note hormones differentially regulate eNOS and NO bioavailability, with estrogen, insulin and growth hormone increasing it, and progesterone and glucocorticoids decreasing it [136]. Antioxidants such as melatonin and Vitamin C have been found to increase blood flow via an NO-dependent mechanism [137]. Evidence is emerging that statins, a class of lipid-lowering drugs, may have differential effects on eNOS and iNOS [138–140]. Of potential interest to chocolate lovers, Taubert et al., reported that regular intake of polyphenol-rich dark chocolate increases NO bioavailability [141]. Treatment of endothelial cells in culture with alcohol increases NO production by enhancing NOS activity [142–144], and alcohol feeding in rats increases aortic NO bioavailability [145]. These NO effects might explain some of the putative cardioprotection seen with moderate alcohol consumption [146–148]. However, while there is a lot of promise in capitalizing on the pre-clinical findings with respect to NO biology, translation to new modalities for cardiovascular disease management has not been uniformly fruitful [134]. Results of dietary nitrate supplementation in clinical trials, for example, have yielded variable outcomes with some beneficial, and some with no measurable effect [149,150].

Fig. 3.

Common risk factors for cardiovascular disease (CVD) such as those listed here are associated with 'endothelial dysfunction', a condition synonymous with reduced biological activity of nitric oxide (NO), and believed to be an initiating factor for atherogenesis.

Moreover, it is important to note that certain conditions such as hypercholesterolemia, hypertension and diabetes can lead to increased vascular production of superoxide (${\mathrm{O}}_2^{-}$). Superoxide can react with the increased NO also produced in such circumstances forming the strong oxidant peroxynitrite (ONOO⁻⁾ that causes protein nitration, lipid peroxidation, DNA damage and cell death [151]. Excessive peroxinitrite is believed an important contributor to cardiovascular pathologies including atherosclerosis. Thus, NO must be considered a pleiotropic molecule, with both physiological and pathological potential as it has the capacity for a dual role as a protective agent, or as a target for generating harmful peroxinitrite, depending on the oxidative milieu [152,151]. As with most things in biology, it is the ’balance’ that is important and that determines physiological homeostasis or pathology development.

10.2. PGI₂

Another important endothelial-derived vasodilator and athe-roprotective agent, whose release is stimulated by shear stress and in response to acetylcholine, is prostacyclin (PGI₂) [153,154]. PGI₂ is synthesized by cyclooxygenase-2 (COX-2) from arachidonic acid [155] and increases the second messenger cyclic adenosine monophosphate (cAMP) in smooth muscle cells as well as in platelets. PGI₂ mediates its actions through specific cell surface, 7-membrane spanning, G protein-coupled receptors, known as PG1₂ receptor or IP [156,157]. In contrast to NO, PGI₂does not appear to contribute to the maintenance of basal vascular tone of large conduit arteries. However, in patients with a decreased NO bioavailability as in atherosclerosis, COX-2-derived prostaglandins can play a compensatory role [158]. Moreover, PGI₂ exerts important platelet inhibitory effects and can act synergistically with NO in this regard [159]. Like NO, PGI₂ can also inhibit leukocyte adhesion and vascular smooth muscle proliferation [160] and crosstalk between the two agents exists. PGI₂ facilitates the release of NO by endothelial cells [161], and conversely, the action of PGI₂ in vascular smooth muscle is potentiated by NO and NO indirectly via cGMP inhibition of phosphodiesterase 3, prolongs the half-life of cAMP [162]. PGI₂'s effects counteract those of thromboxane (TXA2) that is synthesized through COX-1 and causes platelet aggregation, vasoconstriction and vascular proliferation, and the balance of these two prostanoids is thus important in cardiovascular hemostasis [157]. In accordance with the concept of PGI₂ being atheroprotective, an orally active PGI₂ stable analog, Bera-prost, or PGI₂ synthase (PGIS) gene transfer inhibited neointimal formation in animal models of arterial injury [163,164]. Beraprost is widely used in therapy for pulmonary artery hypertension and chronic arterial obstruction [165]. The withdrawal of the Cox-2 inhibitor Vioxx in 2004 due to increased cardiovascular events supports the critical role of PGI₂ in inhibiting atherothrombosis in humans. In a clinical study, the IP variant (R212C), which was defective in cAMP production, was closely linked to disease severity and adverse cardiovascular events in patients with cardiovascular diseases [166]. In addition to the classical IP-cAMP pathway, evidence suggests that PGI₂ also mediates its effects on the vasculature via the nuclear receptor peroxisome proliferator activated receptor δ (PPAR)[167].

10.3. EDHF

Any vasodilator response that persists after combined inhibition of NO (e.g., by L-NMMA) and PGI₂ (e.g., by aspirin) is attributed to substances that cause hyperpolarization of the underlying vascular smooth muscle via a mechanism involving increased potassium (K⁺) conductance resulting in a reduction in intracellular K⁺ and the subsequent depolarization of the cell and relaxation. Molecules and mediators postulated to act as EDHF include K⁺ [168], cytochrome P450 metabolites [169], lipoxygenase products [170], reactive oxygen species (H₂O₂) [171], C-type natriuretic peptide [172], and electrical coupling through myoendothelial gap junctions [173].

11. Vasoconstrictors

Opposing the action of the endothelial-derived vasodilators (NO, PGI₂, EDHFs) are a number of endothelial-derived vasoconstrictors, chief among them endothelin-1 (ET-1), angiotensin-II (ANG-II) and vasoconstrictor prostaglandins.

11.1. ET-1

In 1988, Dr. Yanagisawa's group identified endothelin-1 (ET-1), a 21 amino acid peptide generated by the endothelin converting enzyme (ECE)-mediated cleavage of a larger precursor ‘big endo-thelin-1’ (Big ET-1), as the first endothelium-derived contracting factor [174]. ET-1 is one of the most potent endogenous vasoconstrictors (for review [175]. It is produced mainly by endothelial cells, but also by vascular smooth muscle cells, leukocytes and macrophages [175]. ET-1 and NO are functionally closely interdependent [176]; NO strongly inhibits the release of ET-1 from the endothelium and ET-1 strongly inhibits NO-mediated vasodilation and vice versa [175,177]. There are 3 isoforms of endothelin: ET-1, −2 and −3, with ET-1 most relevant in the vasculature and in contributing to the maintenance of vascular tone [178]. At elevated concentrations, associated with pathological conditions, ET-1 is also pro-inflammatory and promotes smooth muscle proliferation [179,180]. Elevated levels of ET-1 have been implicated in the pathogenesis of hypertension, vasospasm and heart failure, in addition to atherosclerosis. Increased expression of ET-1 and ECE is seen in human arteries at different stages of atherosclerosis [181]. The biological effects of ET-1 are mediated through activation of two receptors, ET_A and ET_B [182,183]. ET_A and ET_B receptors expressed on vascular smooth muscle cells mediate the vasoconstrictor effects of ET-1. The ET_B receptor is also found on endothelial cells, where its activation results in vasodilation mediated by PGI₂ and NO release [184] (Fig. 4). ETB also plays a role in the clearance of circulating ET-1 mediated by ligand-receptor complex internalization and intracellular degradation. Therefore, a reduction in ET_B number, or ET_B receptor blockade, might reduce ET-1 clearance, increasing ET-1 concentrations without altering production.Several ET receptor antagonists have been developed including mixed antagonists Bosetan and the more potent Macitentan, and the selective ET_A antagonist Ambrisentan [185,186]. Restoration of the impaired activity of the NO system is seen following selective ET_A receptor antagonism in a range of animal models of atherosclerosis, together with a reversal of atherosclerotic lesion development [187,188]. However mixed ET_A/B receptor antagonism also seemed beneficial in inhibiting lesion development in mice [189]. Whether or not both receptors should be blocked to provide most clinical benefit in the case of cardiovascular disease is a matter of ongoing debate. More recently, combined inhibitors of ECE and neural endopeptidase (NEP) such as SLV306 (daglutril) are being widely tested in clinical trials [186].

Fig. 4.

Big endothelin-1 (inactive precursor peptide) is cleaved by endothelin converting enzyme (ECE) to form Endothelin-1 (ET-1). Endothelin receptors ET_A and ET_B are expressed on vascular smooth muscle cells (SMC) and mediate vasoconstriction. ET_B is also expressed on endothelial cells (EC) where its activation can result in vasodilation mediated by nitric oxide (NO) and prostacyclin (PGI₂) release.

11.2. Angiotensin-II/AT1R

The renin–angiotensin system (RAS) is a circulating hormonal system that regulates blood pressure, blood flow, fluid volume and electrolyte balance. Recent findings that the RAS is activated locally in the heart and vessel wall have expanded that view and it is now appreciated that it plays a critical role in the initiation and progression of atherosclerosis [190]. Indeed, with the exception of renin (which cleaves angiotensinogen to form angiotensin I), all components of the RAS are produced in the vasculature. Endothelial cells express angiotensin converting enzyme (ACE), the dipep-tidyl carboxypeptidase that converts angiotensin-1 to the physiologically active angiotensin-II (Ang-II) [191]. Both local and circulating Ang-II exert their activities through the binding to Ang-II type 1 (AT1) or type 2 (AT2) receptors. AT1 receptors are widely expressed on different cell types involved in atherogenesis and mediate most of the physiological and pathophysiological actions of Ang-II. A role for the AT2 receptor in cardiovascular disease remains controversial (for review [192]). Ang-II is a proathero-sclerotic mediator; it causes vasoconstriction and regulates not only adhesion molecule (VCAM-1, ICAM-1, P-selectin) expression but also cytokine, chemokine, and growth factor secretion within the arterial wall [193]. ACE inhibitors and angiotensin receptor blockers (ARBs) are established pharmacological tools in both primary and secondary prevention of atherosclerotic cardiovascular disease [194]. The endothelium is the main target organ for many cardiovascular drugs including ACE inhibitors, and evidence shows that their beneficial effects are not only due to blood pressure lowering but also due to a direct anti-inflammatory activity [194].

12. Antithrombotic/prothrombotic balance

The endothelium crucially maintains the lining of the blood vessel as nonthrombogenic [195]. An important pathophysiological consequence of endothelial dysfunction is a disturbed hemostatic balance of thrombotic and fibrinolytic states leading to altered blood fluidity and possible pathologic hemorrhage or clot formation at sites of vascular injury. In regard to its physiological ‘active anti-thrombotic’ role, endothelium associated factors such as the potent inhibitors of platelet aggregation PGI₂ and NO, as well as heparin-like molecules, thrombomodulin, Ecto-ADPase, tissue plasminogen activator and urokinase all may contribute [196]. The endothelium is also capable of active pro-thrombotic behavior; it synthesizes thromboxane A₂ (TxA₂) which promotes platelet aggregation, it expresses adhesive co-factors for platelets such as von Willebrand factor, fibronectin and thrombospondin, and procoa-gulant factors such as factor V [197]. Moreover, the endothelium can be activated by various pathophysiologic stimuli to express tissue factor, a trigger for the fibrin-generating coagulation cascade [198,199]. Endothelial cells can also generate an inhibitor of the fibrinolytic pathway (plasminogen activator inibiyor-1 (PAI-1), which can reduce the rate of fibrin breakdown [200]. Thus, in physiologic conditions, the normal endothelium actively supports the fluid state of flowing blood and prevents activation of circulating cells, whereas injury to the endothelium triggers a series of biochemical and molecular reactions favoring blood arrest and vessel wall repair. In the latter case (i.e., dysfunctional endothelium), platelets may aggregate and stick to the endothelium thereby activating it and encouraging leukocyte adhesion, as well as releasing platelet-derived growth factors that stimulate intimal hyperplasia [201,202].

13. Endothelial progenitor cells

Evidence suggests that the injured endothelial monolayer may be regenerated by circulating bone marrow-derived endothelial progenitor cells (EPC), which accelerate re-endothelialization and protect against the initiation and progression of atherosclerosis [203–205,206]. EPC represent less than 1% of the cells in blood. They are generally purified from blood by density gradient centrifugation for mononuclear cells followed by attachment onto tissue culture plates, and characterized at early stages after attachment by positive expression for both an endothelial marker such as VEGFR2 and for hematopoietic stem cell markers CD133 and CD34. After expansion for several weeks they express endothelial markers including vascular endothelial (VE)-Cadherin and Von Williebrand factor (vWF) [207]. It should be noted, however, that controversy exists with respect to the identification and characterization of ‘EPC’ which may actually encompass several different populations, as well as with regard to their definitive role in the pathophysiology of atherosclerosis vs in angiogenesis [208,209]. Insufficient blood flow resulting in local tissue ischemia is recognized as a potent stimulus for recruitment of EPC [206]; EPC express chemokine receptor type 4 (CXCR-4) receptors that bind stromal-derived factor-1 (SDF-1) that is released at sites of ischemia [210]. Higher circulating levels of progenitor cells reflect greater repair capacity and have been shown to reduce the progression of atherosclerosis [211 ]. Age is a significant risk factor for the development of cardiovascular disease and interestingly, EPCs are subject to age-associated changes that diminish their number in circulation and function [212]. Decreased EPC numbers have also been associated with metabolic syndrome, which confers a high risk for cardiovascular disease [213]. Given the critical role of the endothelium in vascular health and the importance of maintaining a ‘healthy’ endothelium as opposed to a ‘dysfunctional’ one, clinical benefit may be derived from harnessing EPC that can differentiate into mature endothelial cells to use as therapy for vascular disease. With this in mind, strategies that have been employed to enhance EPC number and functionality include statins, ACE inhibitors, angiotensin receptor blockers and peroxisome proliferator-activated receptor gamma (PPAR-γ) agonists, although the mechanisms involved remain unclear [213]. Other clinical approaches involve providing exogenous EPC to the circulation to replenish numbers reduced in diseased patients [214]. Finally, the discovery of ‘resident’ stem/progenitor cells in the vessel wall that can differentiate into all types of vascular cells [215], and evidence that endothelial cells may transform into smooth muscle-like cells [216], has complicated the vascular ‘progenitor cell’ research field even further.

14. MicroRNAs (miRNAs)

Discovered in 1993, microRNAs (miRNAs) are short (19–25 nucleotides long) sequences of non-coding RNA that regulate genes at the post-transcriptional level. It is estimated that there may be up to 5000 miRs that finely tune the expression of a multitude of protein-coding genes [217], including those involved in signaling pathways controlling proliferation, differentiation and apoptosis. It is hardly surprising then, that a burgeoning role for miRNAs in endothelial dysfunction and atherogenesis has been recognized. Indeed, an expanding body of literature indicates that miRNAs affect nearly every aspect of atherosclerotic plaque initiation, development and progression including inflammation/chemo-kines, cholesterol homeostasis, cell adhesion, vascular cell proliferation/apoptosis, and EPC function (recently reviewed in Ref. [218,219]). For example, recent studies report that miR-24 targeting matrix metalloproteinase-14 (MMP-14) regulates macrophages [220], miR-126–5p promotes endothelial proliferation by suppressing Notch1 inhibitor delta-like 1 homolog (Dlk1) [221], miR-126 also inhibits VCAM-1 expression and leukocyte adhesion to endothelial cells [222], miR-145 regulates SMC differentiation and promotes plaque stability [223], miR-155 reduces inflammatory responses and enhances macrophage cholesterol efflux [224], and miR-302a modulates cholesterol homeostasis by affecting the ATP-binding cassette (ABC) transporter ABCA1 [225]. MicroRNAs are actively secreted by endothelial cells under physiological and pathophysiological conditions and can be quantified in blood. These circulating miRNA, which are very stable, are being viewed as promising biomarkers for a number of cardiovascular diseases [226]. Moreover, flow-sensitive miRNAs known as ‘mechano-miRs’ have also been identified in endothelial cells [227]. These include miR-10a, miR-23b, miR-21, miR-663, miR-92a, miR-143/145, miR-101 and miR-126; all of which regulate EC gene expression, affecting EC health or dysfuction and thus, atherogenesis. Given the important role of gene regulating miRs in endothelial function many are currently being considered as potential therapeutic targets to treat atherosclerosis. Therapies might consist of either introducing ‘atheroprotective’ miRs, or depleting ‘atherogenic’ miRs using anti-sense miRNAs.

15. Endothelial-mesenchymal transition (EndoMT)

Vascular endothelial cells can demonstrate plasticity in both developmental and pathological conditions. Generation of mesenchymal cells from endothelium, known as endothelial-mesenchymal transition (EndoMT), is a crucial step in endothelial cell differentiation to several lineages including fibroblasts, mural cells and adipocytes. EndoMT has been implicated in vascular remodeling associated with vein grafting, arteriosclerosis and atherosclerosis [228]. TGF-β has been identified as key in driving EndMT progression [229]. Murine cell lineage–tracing experiments showed that endothelial-derived cells contribute to neointimal formation through EndoMT following vein grafting which is dependent on early activation of the TGF-β Smad2/3-Slug signaling pathway [230]. Antagonism of TGF-β signaling decreased EndoMT and resulted in less neointimal formation when compared to controls. Histological examination of human vein grafts validated the murine data. Furthermore, as caveolin-1 is important for TGF-β receptor internalization and TGF-β signaling, it also appears critical for regulation of EndoMT [231]. EndoMT may be a link between factors initiating atherosclerosis, such as inflammation and disturbed flow, and the tissue remodeling that leads to plaque formation [232]. It appears that oscillatory shear stress reduces fibroblast growth factor receptor 1 (FGFR1) expression and FGF signaling, resulting in activation of endothelial TGF-β signaling and induction of EndoMT. Chen et al., demonstrated that ApoE −/−mice with disrupted ‘protective’ FGF endothelial signaling as a result of cell-specific deletion of FGF receptor substrate 2α exhibited extensive development of EndMT and increased neointima formation [232]. Moreover, in human patients the level of coronary artery disease correlated strongly with loss of endothelial FGFR1 expression, activation of endothelial TGF-β signaling, and the extent of EndoMT [232]. EndoMT is thought to promote plaque growth by increasing deposition of fibronectin and increasing expression of cell adhesion molecules ICAM and VCAM, thereby further promoting recruitment of circulating monocytes and leukocytes and inducing formation of new mesenchymal cells, leading to expansion of neointima. In a separate study, Moonen et al., reported that while steady flow inhibited EndoMT, endothelial cells exposed to disturbed flow underwent EndoMT and exhibited atherogenic differentiation. Gain- and loss-of-function studies established a pivotal role for ERK5 signaling in the inhibition of EndoMT [233]. Collectively, these data suggest that inhibition of EndoMT might prove to be a novel therapeutic strategy to treat atherosclerosis.

16. Summary

The endothelium is critical in maintaining physiological balance in the vasculature and may be considered the gatekeeper of vessel health. Endothelial cells synthesize and release many factors that locally regulate permeability, vascular tone, smooth muscle cell growth and migration, inflammatory responses and platelet function. Upsetting this tightly regulated balance, such as occurs when the endothelium becomes dysfunctional in response to a variety of different risk factors, leads to the initiation and development of atherosclerotic lesions. The endothelium is, thus, an important target for therapies, both current and in the development phase, for cardiovascular disease.