OUP user menu

Redox signalling in vascular responses to shear and stretch

Stephanie Lehoux
DOI: http://dx.doi.org/10.1016/j.cardiores.2006.05.008 269-279 First published online: 15 July 2006

Abstract

Blood vessels are permanently exposed to stretch and shear stress due to blood pressure and blood flow. Significant variations in the mechanical environment, of physiological or pathophysiological nature, occur in vivo. These trigger acute changes in vessel diameter that tend to restore basal levels of tensile and shear stress. However, when altered mechanical conditions persist, they lead to compensatory phenotypical modulation of the endothelial and vascular smooth muscle cells, producing structural and functional modifications of the arterial wall. Such vascular remodelling is a fundamental basis of normal vessel growth and adaptation. However, when the vascular environment changes, due to humoral, metabolic or surgical alterations, for example, mechanical factors may actually exacerbate the underlying conditions and contribute significantly to disease progression. Several studies have demonstrated that reactive oxygen species are induced in the vascular response to changes in shear stress or stretch. It appears that the balance between oxidant and antioxidant generation, which is directly determined by the nature of the mechanical stimulus, can greatly influence the process of vascular remodelling, contributing to both transient and more prolonged adaptations.

Keywords
  • Mechanotransduction
  • Shear stress
  • Stretch
  • Remodelling
  • Oxygen radicals

1. Introduction

Blood vessels evolve in a complex environment where mechanical forces interact to shape and mould the vascular wall. Blood flow creates a frictional force acting in the longitudinal direction at the blood–endothelium interface, expressed as shear stress, which is directly related to the flow-velocity profile. Blood pressure, on the other hand, stretches the vessel wall, generating radial and tangential forces that depend on the pressure applied and the dimensions (diameter and thickness) of the vessel. Mechanical forces are important determinants of vascular homeostasis; laminar shear stress exerts potent anti-apoptotic [1] and anti-atherosclerotic effects [2] on endothelial cells (ECs), whereas a basal level of tangential strain is necessary for maintenance of vascular smooth muscle cells (VSMCs) in a differentiated state [3]. The protective effects of laminar shear stress are lost at arterial branch points and curvatures, where flow is oscillatory or turbulent, increasing the propensity for atherosclerotic plaque formation in such segments. Moreover, when significant variations in mechanical forces occur, they induce structural remodelling in the vascular wall that underlies a trend to re-establish baseline mechanical conditions. Much data has accumulated over the years to implicate reactive oxygen species (ROS) both in the maintenance of steady vessel wall conditions and in the vascular response to altered flow or pressure settings.

2 Mechanical generation of ROS

The basic product of enzymatic ROS production is the superoxide anion (O2), which is transformed quickly into hydrogen peroxide (H2O2) by superoxide dismutase (SOD). The H2O2 is transformed in its turn by two enzymes, catalase and glutathione peroxidase (GPx). Vascular cells comprise diverse sources of ROS, including NAD(P)H oxidase, xanthine oxidase (XO), uncoupled endothelial nitric oxide (NO) synthase (eNOS), cytochrome P450 and the mitochondrial respiratory chain (reviewed in [4,5]) (Fig. 1). ROS production is kept in check by antioxidant defences such as SOD, catalase and GPx, but also thioredoxin (Trx) and heme oxygenase (HO-1). By interfering with the expression and activity of these enzymes, mechanical forces can regulate the redox balance in the vascular wall.

Fig. 1

Oxidant and antioxidant pathways induced by mechanical stresses in vascular cells. Changes in shear stress or strain stimulate superoxide generation through the activation of NAD(P)H oxidase, xanthine oxidase (XO) or the mitochondrial respiratory transport chain. In turn, different reactive oxygen species may be produced as O2 is transformed to H2O2 by superoxide dismutase (SOD) or to peroxynitrite through interaction with NO, generated by the endothelial NO synthase (eNOS). Mechanically induced catalase (CAT) and glutathione peroxidase (GPx) can inactivate H2O2. Moreover, in response to steady shear stress, the Nrf transcription factor may bind the antioxidant response element (ARE), upregulating the expression of GPx-1, NAD(P)H quinone oxidoreductase 1 (NQO1), ferritin, heme oxygenase (HO-1), microsomal epoxide hydrolase (EPHX1), glutathione S-transferase (GST) and gamma-glutamylcysteine synthase (γ-ECS). The double-headed arrow indicates co-dependence between NAD(P)H oxidase and XO expression.

2.1 Induction of ROS by shear stress

Many experiments have shown accrued production of ROS in vascular cells exposed to changes in the flow environment, and in this respect several potential sources of free radicals have been identified, determined in part by the nature of the shear stimulus, laminar (or pulsatile flow with an ample mean shear rate) or oscillatory, but also by the strength and duration of the flow stimulus. Consistent with the notion that the frictional force exerted by blood flow acts essentially on the endothelial surface, most shear stress signalling studies are undertaken using cultured ECs.

Generally, exposure of ECs to oscillatory shear stress is associated with a prolonged ROS generation [6]. In ECs submitted to 4–8 h of oscillatory flow, mRNA expression of both gp91phox (Nox2) and Nox4 subunits of NAD(P)H oxidase was increased, concomitant with a rise in detected ROS [7]. Moreover, whereas Nox4 upregulation was no longer observed at later time points, mRNA levels of gp91phox remained high [8], and expression of both Nox1 and p22phox was enhanced, along with superoxide production [8,9]. Accordingly, oscillatory shear stress induced O2 production in ECs from wild-type mice, but failed to do so in ECs from mice lacking the p47phox subunit of NAD(P)H oxidase [7,8], indicating that this enzyme truly regulates oscillatory shear-induced ROS production. Nevertheless, in bovine aortic ECs, the increase in O2 and H2O2 production associated with oscillatory shear stress coincided with decreased xanthine dehydrogenase (XDH) protein levels and enzymatic activity, tipping the balance in favour of xanthine oxidase, and fittingly it was inhibited by the XO inhibitor oxypurinol [10]. Though these reports may appear to be contradictory, it is not necessarily the case. ECs lacking p47phox exhibit not only dramatically depressed O2 production, but also minimal XO protein and activity. Furthermore, transfection of these cells with p47phox restores XO protein levels, suggesting that NAD(P)H is important for maintaining XO levels [10]. Hence, crosstalk between radical-producing pathways could very well determine the oxidative state in cells.

Unlike oscillatory flow, laminar shear stress induces only a transitory activation of NAD(P)H oxidase in cultured ECs [6,11], and at length both gp91phox and Nox4 mRNA expression are downregulated, with an accompanying reduction in O2 synthesis [7]. This would explain the chronic protective role of physiological values of laminar shear stress. However, data obtained in whole vessels point to a more complicated scheme. On the one hand, in human coronary arteries exposed to an acute increase in shear stress, H2O2 production and flow-induced dilatation were not affected by the NAD(P)H oxidase inhibitor apocynin, but were instead abolished by the mitochondrial complexes I and III inhibitors rotenone and myxothiazol [12], indicating that the source of laminar shear stress-induced ROS may depend on EC culture conditions. On the other hand, abnormally high flow was recently shown to upregulate NAD(P)H oxidase-dependent ROS production chronically in carotid arteries [13], suggesting that supra-physiological levels of shear stress have a protracted influence on the oxidative status. Hence, superoxide generation in ECs at normal laminar shear stress is a transient response to the change in shear rate, whereas ROS production persists in ECs exposed to oscillatory or abnormally high shear stress.

Beyond their effects on ROS-producing enzymes, changes in flow can also induce antioxidant pathways. In ECs exposed to laminar shear stress, the rise in intracellular ROS levels was blocked by either antioxidant N-acetyl-cysteine (NAC), known to increase cellular glutathione and SOD levels, or by catalase [11], showing that antioxidants can effectively counterbalance shear-induced oxidative stress. But more importantly, ECs can produce their own defences against free radicals, as demonstrated in a number of cell culture assays. A short-term exposure to shear stress significantly increased redox regulator thioredoxin-1 (Trx-1) mRNA and protein levels in ECs [14]. Also, both oscillatory and laminar shear stress were shown to induce HO-1 acutely in ECs [6]. However, the protective effect of steady shear stress was prolonged by increased Cu/Zn SOD production [6,15]; as a result, intracellular O2 was higher in cells exposed during 24 h to oscillatory flow than to laminar shear stress [6]. These observations were confirmed in porcine coronary arteries, where Cu/Zn SOD mRNA expression was enhanced after 4 h of high flow [16]. Selective gene upregulation in ECs by sustained laminar but not turbulent or oscillatory shear stress was also shown for Mn SOD [17] and glutathione [18], as well as for genes that contain an antioxidant responsive element (ARE) or ARE-like transcriptional regulatory sequence in their promoters, including glutathione peroxidase (GPx-1) [19], NAD(P)H quinone oxidoreductase 1 (NQO1), HO-1 [20], ferritin (heavy and light chains), microsomal epoxide hydrolase, glutathione S-transferase and gamma-glutamylcysteine synthase [21]. These combined studies explain how long-term oscillatory and steady shear stress produce divergent effects on ROS-producing and ROS-neutralizing enzymes, tipping the balance towards an oxidative state in the former condition and anti-oxidative environment in the latter.

2.2 Induction of ROS by stretch

As could be expected, shear stress is not the only mechanical stimulus that induces an oxidative response in ECs. Several studies have indeed reported that ROS are increased in both ECs and VSMCs exposed to steady or cyclic strain, and again many sources of ROS have been implicated in this response. Interestingly, while applying a 10% cyclic stretch to vascular cells in culture stimulated the production of O2, a cyclic stretch of 6% did not [22,23]. Similarly, in cultured whole vessels, 5% cyclic stretch failed to induce ROS production, whereas 10% and 20% cyclic stretch activated ROS generation equally [24]. It therefore appears that a certain threshold of stretch is necessary and sufficient to activate ROS release.

In cultured ECs, one study showed that oxidant induction by 6-h cyclic strain could be prevented by using the mitochondrial inhibitor rotenone or by rendering the cells mitochondria deficient, whereas inhibitors of NO synthase, xanthine oxidase or NAD(P)H oxidase had no such effect [25]. However, most reports ascribed free radical production by cyclically stretched ECs to NAD(P)H oxidase. Cyclic stretch increased the endothelial mRNA expression of p22phox [26], and ECs treated with the NAD(P)H oxidase inhibitor dipehyleneiodonium (DPI) [22] or carrying a dominant-negative mutant of NAD(P)H oxidase subunit Rac (RacN17) [27] showed reduced superoxide generation in response to cyclic strain. Disparities did arise regarding the kinetics of ROS stimulation by cyclic stretch in ECs; one group found that cyclic stretch up to 6 h increased superoxide production, but that after 24 h of stretch O2 returned to near control levels [22], whereas another showed elevated H2O2 release after both 6 and 24 h of cyclic strain [28]. Experimental differences may lie in the extent of eNOS activation, as long-term NO production was hypothesize to scavenge stretch-induced ROS in the first study [22]. Catalase activity, induced by cyclic stretch [29], may also have dampened ROS generation differentially at length. Ultimately, the antioxidant potential of the stretch stimulus in ECs appears to be less important than that of shear stress, particularly as regards expression of ARE-targeted genes such as GPx-1 [19], which may explain why ROS-generating mechanisms may sometimes prevail in cyclically stretched ECs.

It remains that the mechanical load in vessels is mostly supported by the VSMCs, and appropriately many stretch studies have focused on these cells in culture. In human coronary artery VSMCs exposed to pulsatile stretch, the observed time-and stretch-dependent increased in superoxide production was inhibited by NAD(P)H oxidase inhibitors, but not by xanthine oxidase or cyclooxygenase inhibitors [23]. In agreement, rapid ROS formation in stretched wild-type mouse VSMCs coincided with p47phox membrane translocation, indicative of NAD(P)H oxidase activation, followed by an increase in Nox1 transcripts, whereas ROS formation was completely abrogated in p47phox − / − VSMCs [30]. Superoxide formation due to cyclic strain was counterbalanced by an increase in GPx and glutathione reductase activities, moderating oxidative stress, in rat VSMCs [31].

However, given that the phenotype of VSMCs is altered in cultured cells, being synthetic rather than contractile, and that the response of VSMCs to the mechanical environment is influenced by the extracellular matrix components on which they are cultured as well as their 3-D environment [32], it is noteworthy that a number of studies regarding stretch-induced ROS have been conducted in arteries ex vivo or in vivo. One report established that 24-h cyclic stretch induced superoxide generation in cultured whole aortas, through NAD(P)H oxidase [24]. Subsequent studies also demonstrated increased ROS in vessels exposed to elevated steady stretch rather than cyclic stretch. In both isolated rat femoral branches and bovine coronary arteries, 20–30-min high steady stretch increased the phosphorylation of the NAD(P)H oxidase subunit p47phox [33] and caused its cytosolic to membrane translocation [34], coincident with increased vascular O2 generation. More sustained elevated stretch also induced NAD(P)H oxidase-dependent superoxide production [35,36]. In an in vivo model of aortic banding, expression of the NOX1, gp91phox, p47phox and p67phox subunits of NAD(P)H oxidase, but also of Cu/Zn SOD and Mn SOD, were found to be markedly increased in the hypertensive aortic segment above the coarctation, whereas levels of these proteins were virtually unchanged in the segment below coarctation, where blood pressure is not elevated [37,38]. Hence, increased levels of superoxide and superoxide-producing enzymes, particularly NAD(P)H, are observed in stretched vascular cells and whole vessels, and it appears that the antioxidant response is not sufficient to overcome the effects of ROS-producing enzymes in this context.

2.3 Pathways of mechanical induction of ROS

Based on the localisation of its NOX subunits, anchored in the cell membrane, it is conceivable that NAD(P)H oxidase could be directly induced by changes in the mechanical environment of vascular cells. Aside from this, several alternative pathways of ROS induction by strain have been evoked. In different studies, oscillatory stretch-induced O2 production in ECs was blocked by noggin, a bone morphogenic protein (BMP) antagonist [8], or by protein kinase C (PKC) inhibition [39]. Furthermore, the endothelial oxidative response to stretch was found to depend on EC interaction with matrix proteins, being sensitive to RGD peptides or to α2 and β integrin-blocking antibodies [40]. A separate ROS production pathway was proposed in cultured VSMCs, involving transforming growth factor TGF-β1 in NAD(P)H oxidase activation by cyclic stretch [41]. Unlike cultured ECs, in whole arteries, PKC did not mediate the generation of O2 in response to cyclic stretch, but rather intervened in the oxidative response to steady high intraluminal pressure [24,33]. PKC inhibitors also reduced the accrued production of O2 detected in hypertensive forelimb arteries of rats with aortic banding, compared with normotensive hindlimb arteries [35]. Differences in the matrix protein environment could explain the disparity in PKC sensitivity of ROS production between cultured cells and whole vessels exposed to cyclic stretch. Moreover, distinct oxidative (or antioxidant) pathways may be solicited according to the type, strength or duration of the mechanical stimulus, regulating ROS through different enzymes (Fig. 2).

Fig. 2

Oxidant and antioxidant mediators involved in the alteration of vessels exposed to mechanical strain. Effects of intraluminal stretch and oscillatory shear stress, inducing the MAP kinases JNK, p38 and ERK1/2, or NF-κB, are mostly mediated by superoxide. In comparison, generation of reactive oxygen species in response to laminar shear stress is offset by NO and antioxidant production, countering the activation of MAP kinases and NF-κB. Changes in the mechanical environment invariably induce vascular remodelling, and both NO and ROS can contribute to this process. On the other hand, modulation of NF-κB activity by laminar shear stress opposes the pro-atherosclerotic effects of stretch or oscillatory shear stress, whereas stretch impedes flow-dependent dilatation.

3 Roles of mechanically induced ROS

Recent data suggest that oxygen free radicals, as well as endogenous antioxidants, probably have critical signalling functions in cells [42]. Numerous studies have implicated O2 and H2O2 in different aspects of the vascular response to stretch or shear stress, and the breakdown products of H2O2, including lipid hydroperoxides, are also biologically active. As a whole, oxygen free radicals thus comprise several potential second messengers and are thus poised to influence mechanosensitive cell signalling but also vascular remodelling as a whole.

3.1 ROS-dependent responses to acute changes in shear stress or pressure

Alterations in stretch or shear stress invariably produce transformations in the vessel wall that will aim to accommodate the new conditions and to ultimately restore basal levels of tensile stress and shear stress. Vessels are able to react quickly to any variations in local flow or arterial pressure environments, and adjust their diameters accordingly. Changes in shear, sensed principally by endothelial cells, are countered by the release of vasoactive agonists. Changes in pressure, on the other hand, elicit different responses according to the size of the artery concerned. Conductance arteries expand when exposed to an increase in pressure, whereas in resistance arteries the distending effect of pressure is offset by the development of myogenic tone. Several studies have implicated ROS in these processes.

Flow triggers the production of both NO and ROS, and the interaction between these two components is a key determinant of the ensuing vascular response. In human resistance arteries, preventing ROS generation [12] or scavenging H2O2 with catalase [43] prevented flow-dependent dilatation. Similarly, the increase in eNOS expression associated with exercise training, known to augment endothelial shear stress and oxidative stress, was abolished in vessels of transgenic mice expressing high levels of catalase [44]. ROS therefore appear to regulate NO production and bioavailability in changing flow conditions. Presumably, when shear stress is laminar, NO production is sufficiently elevated to counterbalance ROS generation and bring about vasodilation, whereas during oscillating flow, NO may be quenched by excessive O2, suppressing vasodilation [45].

The balance between ROS and NO also comes into play in arteries exposed to high pressure. A 30-min exposure of arteries to high intraluminal pressure significantly increased vascular O2 production and impaired flow-dependent dilation [33]. Furthermore, superoxide dismutase and catalase, but not catalase alone, prevented the reduction in flow-dependent dilation of arterioles to flow after pressure treatment, indicating that it was the release of superoxide that interfered with NO [46]. In rings from DOCA-salt hypertensive rats, characterized by high basal levels of superoxide, administration of SOD or apocynin completely abolished the development of spontaneous tone in endothelium-intact aortic rings but not in endothelium-denuded or in eNOS inhibitor-treated rings, despite the fact that SOD and apocynin decreased the generation of O2 in all rings [47]. Hence, it appears that high pressure-dependent release of ROS counteracts endothelial NO, which otherwise contributes to flow-dependent dilation and tempers the development of myogenic tone in arterioles.

3.2 ROS-dependent activation of signalling pathways

The mitogen-activated protein (MAP) kinase cascade is an important pathway whereby signals originating from mechanical forces can lead to gene expression and protein synthesis [48]. This pathway implicates the sequential phosphorylation and activation of the cytoplasmic protein kinases MEKK (MAP kinase kinase kinase), MEK (MAP kinase kinase), and finally MAP kinase. The MAP kinase cascade comprises in reality several different pathways, which are triggered in response to various stimuli and initiate distinct cellular responses. The phosphorylation of one of the MAP kinases, which lies downstream from Raf and is present under two isoforms, extracellular signal-regulated kinase (ERK) 1 and 2, leads to the activation of regulatory proteins in the cytoplasm and the nucleus. Other MAP kinases, called stress-activated protein kinases (SAPK) because they are activated by stimuli such as UV light, heat shock, hypoxia or hyperosmolarity, include c-jun amino-terminal (JNK) kinases and p38.

Acute exposure of endothelial cells to laminar shear stress has been associated with oxidant-dependent phosphorylation of both ERK1/2 and JNK. ROS were demonstrated to be associated with temporal gradients in shear-induced ERK1/2 phosphorylation, and treating sheared endothelial cells with the antioxidants NAC or pyrrolidine dithiocarbamate (PDTC) inhibited the ERK1/2 activation [49,50] and reduced the expression of the downstream target c-fos [11]. Similarly, apocynin treatment and the transient overexpression of Cu/Zn SOD abated shear-dependent activation of JNK, pointing at the causative role of NAD(P)H-derived O2 [51]. Complementary studies identified peroxynitrite, the product of NO and O2 interaction, as the mediator in shear stress-induced ERK1/2 and JNK activation [49,51]. Interestingly, not all studies demonstrated increased JNK activity in endothelial cells exposed to shear stress. Moreover, preconditioning of cells with shear stress was found to actually suppress JNK activation by H2O2 through upregulation of glutathione peroxidase [52]. Hence, this signalling pathway is likely to be influenced by the cell culture conditions on top of the shear and ROS environment.

In cultured VSMCs stimulated with cyclic strain, the rapid increase in intracellular ROS was also found to coincide with phosphorylation of ERK1/2, JNK and p38 MAPK, through activation of the NAD(P)H oxidase [53]. Likewise, both steady stretch and chronic cyclic stretch were associated with NAD(P)H oxidase-dependent ERK1/2 and p38 pathway induction in whole arteries [24,34,36]. Actually, O2 levels and ERK1/2 activity were equivalent in freshly isolated arteries and in arteries exposed to chronic cyclic stretch, suggesting that a baseline level of ROS may be persistently generated in conductance vessels, which are naturally exposed to continuous cyclic stretch, maintaining a physiological level of ERK1/2 activity [24].

But MAP kinase pathways may not account for all changes observed in vessels exposed to a change in the mechanical environment, and in this context a role for the transcription factor NF-κB (nuclear factor-κB) has also been evoked. Low shear stress induced NF-κB in ECs via NAD(P)H oxidase [54], and even in VSMCs shear stress activated the redox-sensitive nuclear translocation of NF-κB [55]. The mobilization of NF-κB in both ECs and VSMCs exposed to cyclic stretch was responsive to NAD(P)H oxidase inhibition as well [23,26]. This latter observation was recently validated in the whole vessels exposed to steady high intraluminal pressure [36,56].

Thus far, some ROS-dependent intermediates have been identified in the mechanical activation of the MAP kinase and NF-κB signalling pathways. Exposure of ECs to flow was associated with a rapid, NAC-sensitive phosphorylation of the proline-rich tyrosine kinase PYK2 and p130 Crk-associated substrate (Cas), two kinases known to activate ERK1/2 [57]. ROS-dependent Src and PYK2 activation was reported in stretched ECs as well [39]. Recent studies also indicated that activation of the epidermal growth factor receptor (EGFR) by ROS could participate in the activation of ERK1/2 initiated by cyclic stretch in cultured smooth muscle cells [58] or by acute steady stretch in coronary arteries [34]. This pathway was also recently found to mediate chronic strain-dependent induction of NF-κB [36].

3.3 Responses to chronic changes in shear stress or pressure

Significant variations in mechanical forces, of physiological or physiopathological nature, occur in vivo. When these cannot be countered by acute regulation of vessel wall diameter, they lead to phenotypical modulation of the ECs and VSMCs, producing structural modifications of the arterial wall. In all the cases, vascular remodelling can be allotted to a modification of the tensional strain or shear, and underlie a trend to re-establish baseline mechanical conditions. As such, vascular remodelling is a fundamental basis of normal vessel growth and adaptation. However, when the vascular environment changes, due to humoral, metabolic or surgical alterations, for example, mechanical factors may actually exacerbate the underlying conditions and contribute significantly to disease progression.

3.3.1 Flow-induced remodelling

The most spectacular illustration of the phenomenon of flow-dependent growth is the arteriovenous fistula model (AVF). In carotid artery-jugular vein AVFs, the flow rate in the developing carotids can be multiplied by a factor of up to 8. As long as there is no excessive increase in blood flow, however, shear stress is normalized by a compensatory increase in carotid diameter [59]. One of the striking characteristics of the arterial wall proximal to an AVF is extensive tearing and fragmentation of the internal elastic lamina (IEL) [59,60], which augments arterial distensibility, leading to enhanced vessel diameter. Matrix metalloproteinases (MMPs) are likely instigators of IEL degradation in the vessel wall. MMP-2 and MMP-9 are upregulated shortly after AVF construction, and heightened activity of these enzymes persists until shear stress is normalized [61,62]. Furthermore, several studies have reported that MMP inhibition diminishes flow-mediated arterial enlargement in rat [61,63], rabbit [62] and mouse [13] AVF models, and prevents elastin degradation associated with AVF-induced arterial enlargement [62]. The contribution of NO in flow-induced vascular remodelling has also been well established [13,59,62,64]. After effective long-term blockade of NO synthesis, or in eNOS-deficient animals, artery enlargement and the corresponding wall tissue growth expected in response to increased blood flow are dampened, and the shear stress in flow-loaded common carotid arteries increases proportionately to blood flow.

Recently, the role of ROS in high flow-induced vascular remodelling was demonstrated in a mouse model of AVF [13]. ROS production was enhanced in arteries exposed to chronic high flow, both 1 and 3 weeks after opening of the AVF. NAD(P)H oxidase comprising the p47phox subunit was identified as the major generator of shear stress-induced ROS in the AVF vascular wall; gp91phox, in comparison, played a negligible role in this process. Generation of ROS, as well as NO derived from eNOS activation, led to MMP activation in AVF and modulated flow-induced vascular enlargement. Observations made in p47phox − / − mice and eNOS − / − mice also provided direct evidence that endogenous ROS and NO modulate not only the activity of MMPs, but also their production, in arteries exposed to elevated blood flow. This study showed that long-term structural adaptation to altered blood flow is mediated by ROS.

3.3.2 Pressure-induced remodelling

The effects of mechanical tensile stress on the arterial wall have been extensively described and have been applied to the understanding of hypertension. Tensile stress is a strong determinant of the vascular structure among other factors including sympathetic activity and autocrine and paracrine factors. In hypertension, VSMCs undergo hypertrophy/hyperplasia and synthesize key extracellular matrix proteins in order to increase wall thickness and to normalize tensile stress, resulting in increased arterial wall thickness and rigidity [48,65]. Such structural and functional alterations to both large arteries and arterioles are directly involved in the cardiovascular morbidity and mortality associated with hypertension [66].

Accumulation of collagen is likely to contribute to increased vascular stiffness characteristic of the hypertensive artery. However, in young humans with mild to moderate hypertension, increased collagen deposition is actually associated with reduced vessel stiffness [67], which might reflect an early adaptative phase that paves the way to later vessel stiffness. A recent demonstration that high intraluminal pressure-induced MMP-9 in carotid arteries contributes to increased vessel distensibility points to a role for matrix degrading enzymes in the early stages of hypertensive vascular remodelling [68]. Such stretch-induced MMP activation could be attributed to ROS. Indeed, gelatinolytic activity of pro-MMP-2 and upregulation of MMP-2 mRNA in response to cyclic stretch was abrogated in VSMCs from p47phox − / − mice, compared with cells from wild-type animals [30]. Moreover, ROS-dependent NF-κB activation, as shown in ex vivo and in vivo models of hypertension [33,35], could regulate the transcription of numerous MMPs including MMP-1, MMP-2, MMP-3 and MMP-9 [69–71]. Proliferation of VSMC in hypertension, which requires ERK1/2 [72], may also involve ROS. Either way, it is clear that ROS generation is enhanced in hypertension, exemplified by greater superoxide anion production in aortas from spontaneously hypertensive rats (SHR) [73] and DOCA-salt hypertensive rats [47], than in aortas from normotensive animals. Moreover, the critical contribution of ROS to the evolution of hypertension is substantiated by the finding that treatment with the O2 scavengers or with apocynin for 3 weeks both abolished the increases in O2 generation and prevented the development of hypertension in DOCA-salt rats [47].

3.3.3 Remodelling under changing mechanical conditions

Vein graft disease, a common affliction of veins used as arterial bypass conduits, is characterized by excessive VSMC hyperplasia that brings about occlusive intimal thickening. VSMCs acquire a synthetic phenotype and migrate through the vessel wall, proliferating in the sub-endothelial space [74,75]. In comparison, arterial bypass grafts are much less prone to remodelling. It has long been recognized that altered mechanical factors contribute to vein graft proliferation; pulsatile stretch was found to induce VSMC hyperplasia in saphenous veins but not internal mammary arteries cultured in identical conditions [76], and even venous VSMC exposed to cyclic stretch proliferate more than arterial VSMC [77]. The differences between artery and vein grafts also coincide with greater superoxide production in the latter [78], which can be attributed to enhanced NAD(P)H oxidase subunit expression and activity [79].

Cell proliferation in stretched vein grafts is associated with MMP-2 secretion [80], an observation consistent with the notion that phenotypic modulation of VSMCs occurs when interaction with extracellular matrix proteins is diminished [75]. Indeed, local TIMP-2 gene transfer significantly reduces vein graft remodelling to an artery-like vessel via inhibition of matrix metalloproteinase activity [81]. MMP activation in vein grafts under arterial conditions occurs simultaneously with superoxide production, suggesting a role of ROS in MMP processing [82]. Initial cell death might also drive the proliferative response in vein grafts [83], and indeed the rate of apoptosis is greatly reduced in sheathed grafts, which do not tend to proliferate [84]. Nonetheless, vein graft hyperplasia was found to be doubled in mice lacking the proapoptotic factor p53, but this was likely due to the concomitant MMP-2 upregulation [85]. On the other hand, cyclic strain activates pro-survival pathways such as Akt [86] and NF-κB [23] in cultured VSMC, and intraoperative transfection of the NF-κB decoy into the vein graft wall attenuates of neointima formation [87]. Hence, apoptosis may trigger the onset of vein graft disease, but at length it helps constrain the remodelling response.

3.3.4 Influence of mechanical factors on atherosclerosis

The association between atherosclerotic plaque localization and regional flow patterns is well established. Plaques tend to form in regions of low, disturbed or oscillating flow, such as arterial branch points, bifurcations or in inner curvatures, whereas vessel segments exposed to steady laminar shear stress tend to be relatively lesion-free [88]. NO synthesis in areas of high wall shear stress appears to be a key mediator of atheroprotection, reducing endothelial permeability, migration of leukocytes and VSMC proliferation while simultaneously promoting EC survival [1,17,89]. Prolonged shear stress also prevents inflammatory mediator-stimulated MAP kinase induction and vascular cell adhesion molecule (VCAM-1) expression [90]. Inhibition of apoptosis through upregulation of glutathione [91], Cu/Zn SOD and NO synthase [92] probably plays a protective role in ECs exposed to laminar shear stress as well. Conversely, oscillatory shear stress induces the expression of endothelial adhesion molecules, chemokines and growth factors [93,94] that are important for leukocyte recruitment and extravasation, and is associated with a shift in the balance between NO and ROS levels favouring the latter [6,9,95,96], together with enhanced NF-κB activity [54]. Fittingly, the NF-κB pathway is induced in vascular regions with a high probability for atherosclerotic lesion formation [97]. Moreover, in ECs from wild-type mice, 18-h oscillatory shear stress increases both O2 and monocyte binding, whereas laminar shear stress has the opposite effects. In ECs from p47phox − / − mice, however, O2 production and monocyte binding are no longer induced by oscillatory shear [7]. Along the same lines, ROS scavengers, or knocking down nox1 with the small interfering RNA, blocks intracellular adhesion molecule (ICAM-1) expression and monocyte adhesion induced by oscillatory shear stress in ECs [8].

Mechanical strain characteristic of hypertension may also influence atherosclerotic plaque development. Indeed, carotid artery intima media thickness, an index of atheroma size, is highly correlated with blood pressure in patients [98], and hypertension increases plaque formation in many animal models of atherosclerosis [99–102]. In ApoE − / − mice, creation of a coarctation facilitates plaque formation in the segments upstream of the stenosis, where blood pressure is elevated [103], and similarly in rats with aortic banding, increased expression of adhesion molecules is observed only in segments at high blood pressure [104]. Moreover, levels of soluble cell adhesion molecules are elevated in patients with hypertension, and in both normotensive and hypertensive subjects, systolic blood pressure correlates significantly with soluble ICAM-1 and VCAM-1 [105]. Wall stretch associated with hypertension may well enhance plaque formation through the production of ROS and consequential activation of pro-inflammatory genes [99], a processes that probably implicates NF-κB. Indeed, in ECs NF-κB activation and VCAM-1 mRNA expression during strain were prevented by antioxidants [25]. Treatment with antioxidants NAC or catalase inhibited cyclic strain-induced O2 generation and ICAM-1 mRNA levels, leading to decreased ICAM-1 expression on EC surfaces compared with untreated cells [106]. Cyclic strain-induced monocyte chemoattractant protein (MCP-1) mRNA levels could likewise be inhibited by treating ECs with NAC or catalase [107] or infecting ECs with RacN17, rendering them NAD(P)H oxidase deficient [27].

4 Conclusions

In summary, numerous studies have demonstrated that changes in the mechanical environment disrupt the oxidative balance in vascular cells. ROS tend to be elevated in ECs exposed to oscillatory shear stress or to a transient increase in laminar shear stress, whereas prolonged laminar shear stress is rather associated with a predominantly antioxidant response, unless the shear stress is abnormally high. In vascular cells, both cyclic and steady stretch favour the upregulation of ROS-producing enzymes, although different mechanisms may be involved in the responses to these two stimuli. ROS have been implicated in the acute and prolonged effects of changes in mechanical stresses, regulating immediate responses such as flow-dependent dilatation, but also signalling cascade induction and chronic processes related to vascular remodelling, both physiological and physiopathological. These combined data highlight the important role of ROS in the vascular wall and identify ROS as a potential target in preventing adverse effects in vascular segments exposed to oscillatory shear stress or hypertension.

Footnotes

  • Time for primary review 21 days

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
  55. [55]
  56. [56]
  57. [57]
  58. [58]
  59. [59]
  60. [60]
  61. [61]
  62. [62]
  63. [63]
  64. [64]
  65. [65]
  66. [66]
  67. [67]
  68. [68]
  69. [69]
  70. [70]
  71. [71]
  72. [72]
  73. [73]
  74. [74]
  75. [75]
  76. [76]
  77. [77]
  78. [78]
  79. [79]
  80. [80]
  81. [81]
  82. [82]
  83. [83]
  84. [84]
  85. [85]
  86. [86]
  87. [87]
  88. [88]
  89. [89]
  90. [90]
  91. [91]
  92. [92]
  93. [93]
  94. [94]
  95. [95]
  96. [96]
  97. [97]
  98. [98]
  99. [99]
  100. [100]
  101. [101]
  102. [102]
  103. [103]
  104. [104]
  105. [105]
  106. [106]
  107. [107]
View Abstract