Spotlight on microcirculation: an update

Time for primary review 30 days. Over the past 3 decades there has been an increasing interest in the better understanding of the role of microcirculation under a variety of conditions. Therefore, in 1996 Cardiovascular Research devoted the whole October issue to microcirculation. Two years later the interest in this field is far from being reduced, as the number of articles published has remained high, with over 800 papers per year. The purpose of this article is to review several major studies which have appeared over the last 2 years in this field, which update the various key areas of research that were extensively reviewed in the 1996 issue. Oxygen delivery to tissue has long been considered to take place almost exclusively at the capillary level. However, it has progressively become appreciated that tissue oxygenation is the result of a complex process in which a substantial amount of oxygen is exchanged through arterioles, which in some tissues may be a greater oxygen source than capillaries (see review article by Intaglietta et al. [1]). Very recently, gas diffusion in postcapillary venules has also been supposed to occur, and it has been hypothesized that gas diffusive transfer may play an important role in regulating tissue oxygenation. Given the parallel arrangement of small arteries and veins in several tissues, oxygen transfer from arterial to venous vessels may result in diffusion shunting of oxygen, which may be detrimental to tissue oxygenation; on the other hand, diffusive transfer from arterioles to capillaries and among capillaries might contribute to homogeneous tissue oxygenation. Diffusion of CO2 might also improve tissue oxygenation, since CO2 diffusion from venules to arterioles would reduce pH in arterial blood and thus increase O2 release from hemoglobin. In this regard, it has recently been shown that in rat skeletal muscle diffusion shunting of oxygen from arterioles to postcapillary venules is enhanced during hyperoxia, while during hypoxia CO2 accumulates in peripheral vessels [2]. Thus, diffusive shunting of oxygen and counter diffusion and accumulation of CO2 may represent an important mechanism contributing to the homeostasis of tissue oxygenation levels [2]. Analysis of the spatial distribution of PO2 in venular structures has indicated that longitudinal O2 gradients and spatial heterogeneity are only weakly dependent on mean systemic blood pressure, further supporting the hypothesis that venules may play an important role in regulating oxygen delivery [3]. A recent study in dog intestine has shown that an increase in the vasoconstricting tone causes a redistribution of flow towards mucosa, both under high and low flow conditions, which correlates with the improvement in oxygen extraction ratio [4]. Interestingly, capillary transit time heterogeneity remained unchanged, suggesting that this variable is tightly regulated in the intestine. The long-standing controversy over the determinants of macromolecule transport through the endothelium of microvessels was discussed by Michiel [5]. In addition to convective transport through pores in endothelial cells, at least four different mechanisms of vesicular transport across endothelium have been hypothesized. Evidence available does not allow to clearly eliminate any of them, and it is likely that no single mechanism can account for transport of all macromolecules in all endothelia [5]. Recent evidence in the eel swim bladder has shown that transcytosis of insulin and albumin occurs via different sets of plasmalemmal vesicles, probably through receptor-mediated mechanisms [6]. Much work has focused on the mechanisms of the increase in vascular permeability induced by vascular endothelial growth factor (VEGF), an endothelial cell mitogen that plays a major role in angiogenesis. Angiogenesis is preceded by enhanced microvascular permeability, that is mediated by the interaction of VEGF with the R2 receptor subtype, and implicates nitric oxide and prostacyclin production [7]. Vessel arrangements and biophysical behavior of blood components within the vasculature are of major importance in regulating the transport and exchange function of microvessels (see review by Pries et al. [8]). Changes in rheological properties of the erythrocytes affect resistance to flow; this phenomenon is partly mediated by changes in the thickness of the marginal cell-free layer in microvessels (≤40 μm). This phenomenon has been studied both in vivo and in artificial models of circulation, although it has now become appreciated that microvascular resistance to flow in vivo is higher than in glass tubes of similar diameters [8]. In addition, since microvessels are arranged as a network, vascular function is strongly influenced by vessel architecture. Recent studies have shown that reduced elasticity of microvessel walls can significantly alter the adaptive response to changes in rheological properties, resulting in increased resistance to flow [9]. Analysis of red cell motion through cylindrical micropores has shown that filterability of erythrocytes is dependent on their resistance to transient deformations [10]. Although increased red cell resistance to deformation may impair microcirculation at low temperatures, this phenomenon is not likely to result in significant worsening of flow reduction in vivo, with effects comparable to those resulting from changes in plasma and blood viscosity [11]. The various mechanisms that induce vasodilation in the coronary bed act coordinately and in an integrated fashion to regulate blood flow, with complex interactions [12]. Microvessels of different diameters exhibit different sensitivity to vasoactive stimuli, with a longitudinal gradient in vasoactive responses; for example, response to metabolic vasodilation mainly involves small arterioles (≤40 μm). However, vasodilation of a segment of the microvascular tree affects pressure and flow in both proximal and distal vascular beds, thus eliciting myogenic and shear-stress dependent responses that make an important contribution to the final net effect on vascular resistance and flow. This concept has found further support in a study by Liao and Kuo who, using a modeling approach have shown that about 20% of the adenosine-induced increase in flow is actually dependent upon shear-sensitive mechanisms [13]. In fact, dilation of small arteries in response to adenosine increases flow in spite of decreasing pressure; increased flow in turn activates the shear-sensitive mechanism in up-stream arterioles, further enhancing flow [13]. The complexity of flow regulation is also depicted in a study by Komaru et al., showing that, while activation of G-proteins results in dilation of both large (>130 μm) and small (<130 μm) microvessels, the mechanisms underlying these effects are different and vessel size dependent. In small vessels vasodilation is mediated by activation of ATP-sensitive K+ channels, while in large microvessels it is induced through activation of the nitric oxide pathway and ATP-sensitive K+ channels in a synergistic fashion [14]. These observations further confirm the complexity and intricate regulation of flow in coronary microcirculation. It has recently been appreciated that, in addition to nitric oxide, another gaseous molecule, carbon monoxide, may serve as mediator of vascular cell relaxation, by activating soluble guanylate cyclase to produce cGMP [15]. Carbon monoxide is generated by two forms of heme oxygenase, an inducible enzyme (HO-1), and a constitutive enzyme (HO-2). In the liver, HO-1 has been found in Kupfer cells only, while the constitutive form is present in parenchymal cells, and constitutive release of carbon monoxide by HO-2 appears to play a major role in regulation of microvascular tone, since administration of carbon monoxide-trapping agents results in marked sinusoidal constriction [16]. Interestingly, expression of HO-1 in liver can be stimulated by nitric oxide donors, which induce a progressive increase of HO-1 mRNA and protein activity [17]. An increasing number of observations suggests a role for carbon monoxide as a signalling molecule, in liver and in other organs. Release of carbon monoxide appears to play a major role also in the hemodynamic alterations of endo-toxic shock. In the postischemic heart, increased mRNA levels of heme oxygenase have been reported [18], and it has been suggested that myocardial preservation by nitric oxide may be modulated, at least in part, by carbon monoxide signaling [18]. Both forms of heme oxygenase are expressed in brain [19], where endogenous carbon monoxide is involved in the control of oxytocin release [20]. Finally, induction of HO-1 is also involved in nitric oxide-stimulated keratinocyte proliferation [21]. Taken together, these studies suggest that carbon monoxide might play an important role as a messenger molecule in a number of pathophysiological conditions. The role of pericytes in vessel architecture has also received increasing attention. Pericytes have different morphology and distribution in various tissues, suggesting differences in function [22]. They express contractile protein and may contract, changing capillary resistance to flow, and it has been suggested that pericytes may further differentiate into smooth muscle cells. Communication between pericytes and endothelium involves release of soluble mediators and direct interaction via membrane proteins. Pericytes are now considered important modulators of physiological events like changes in capillary resistance and angiogenesis, and they have also been involved in pathological conditions like hypertension, diabetes and tumor vascularization. The enzyme aminopeptidase A, which is associated with microvessels of all organs in animals, has been found on cell membranes of activated pericytes in conditions associated with neovascularization, while it is present at very low concentrations in normal vessels [23]. This observation supports a regulatory role for pericytes during neovascularization, and may represent a marker of pericytes activation. Evidence supporting a role for pericytes in diabetic retinopathy is discussed below. In the heart postischemic reperfusion is accompanied by an inflammatory reaction, with release of proinflammatory and fibrogenic mediators that promote tissue healing; however, this phenomenon may also result in leukocyte-mediated cardiac injury [24–26]. In fact, activated neutrophils may infiltrate postischemic myocardium through sequential steps of rolling along vessel walls, adhesion to endothelium and migration in tissues, where they release cytotoxic mediators with subsequent myocardial injury [27]. Since reperfusion is accompanied by a decrease in venular shear forces, and by an increase in neutrophil rolling, adhesion and extravasation, it has been suggested that conditions of slow flow may contribute to neutrophil recruitment in tissues. In postischemic mesenteric microvasculature, maintaining shear forces at control values did not affect the number of rolling leukocytes, but it did reduce the number of adherent cells and subsequent microvascular dysfunction [28]. This observation suggests that decreased shear stress does not contribute to initial neutrophil activation, but it is an essential permissive component for neutrophil firm adhesion and extravasation in postischemic microvasculature [28]. Also, in postischemic hearts low-flow reperfusion further increases neutrophil accumulation in coronary microcirculation [29]. These findings may have important clinical implications, since neutrophil accumulation in postcapillary vessels is the major determinant to the no-reflow phenomenon [30]. Clinical studies indicate that the no-reflow phenomenon does occur in man [31]. The occurrence of no-reflow has major prognostic implications, since it is associated with worse contractile recovery, higher incidence of arrhythmias and pericardial effusion, and with late development of left ventricular dilatation and congestive heart failure [31]. Interestingly, no-reflow mostly occurred in all patients with slow flow in the epicardial artery after percutaneous transluminal coronary angioplasty, while it was only present in a minority of patients with almost normal flow in the epicardial artery [32]. These clinical findings might be explained by recent data of Kubes [28] and Ritter and McDonogh [29], which indicate that neutrophil accumulation in postischemic tissues is greatly enhanced under conditions of slow flow. Mast cells may release several proinflammatory mediators, and their activation has been proposed to play a major role in a variety of pathophysiological situations by promoting leukocyte recruitment and inflammation [33]. Mast cell activation has also been involved in the inflammatory reaction that occurs during postischemic reperfusion in the heart [27,34]. Monocytes have also been shown to play an important role in tissue healing, and their recruitment is mediated by different factors. Release of the complement factor C5a is the major product of monocyte chemotaxis in the first hour after reperfusion, while transforming growth factor 1 (TGF-β 1) is released between 60 and 180 min; after 180 min TGF-β 1 and monocyte chemoattractant protein-1 coordinately promote monocyte recruitment [35]. Exposure to oxidized LDL also induces mast cell activation and leukocyte–endothelial cell adhesion [36], events that have been proposed as initial steps in the development of atherosclerotic plaque. Mast cell stabilization attenuated leukocyte–endothelial cell adhesion in response to oxidized LDL, indicating that these cells are involved in the microvascular dysfunction induced by oxidized LDL [36]. Other studies have investigated the mechanisms of neutrophil adhesion to the endothelium. The ability of leukocytes to cross the vessel wall is fundamental, since it allows these cells to reach the site of tissue inflammation where they may exert their actions, and firm adherence to endothelium is the first step of this process [37]. Exposure of CD11 integrins on neutrophil surface plays an important role in this phenomenon, and it has been appreciated that specific integrins may have different roles in cell adhesion. CD11b integrin plays a critical role in mediating binding of neutrophils to fibrinogen and neutrophil degranulation, but it is not necessary for effective neutrophil emigration, which is more dependent upon CD11a [38]. To evaluate the role of P-selectin and ICAM-1 in mediating leukocyte rolling, genetically engineered mice lacking one or both these adhesion molecules have been subjected to trauma- and cytokine-induced inflammation [39]. In normal mice, leukocyte rolling is largely mediated by P-selectin, and in part by ICAM-1. In mice lacking both molecules, leukocyte rolling during inflammation still occurred to some extent, and it was abolished by antibodies directed against E-selectin, clearly demonstrating E-selectin dependent rolling in vivo [39]. Interestingly, leukocyte rolling induced by Tumor Necrosis Factor α is increased by blockade of endogenous nitric oxide production, while it is not affected by exogenous nitric oxide donors [40]. This seems to be a specific phenomenon, since inhibition of nitric oxide synthase did not affect leukocyte recruitment during ischemia–reperfusion [40]. Thus, increased production of nitric oxide seems to play an important role in dampening leukocyte recruitment in response to pro-inflammatory cytokines, while during postischemic reperfusion other mechanisms probably overcome this regulation. It has also been demonstrated that neutrophil migration across Interleukin-1 stimulated cultured endothelium occurs preferentially where the borders of three endothelial cells intersect, independently of tight junctions [41]. Postischemic reperfusion is accompanied by profound endothelial dysfunction, mediated by oxygen radical generation and inhibition of nitric oxide synthase activity [25,42,43]. Reduced nitric oxide production results in upregulation of cell adhesion molecules that in turn stimulates neutrophil recruitment and myocardial injury during reperfusion [25,42,43]. This phenomenon has now been reported in the lung, in which inhaled nitric oxide attenuated the microvascular leak induced by ischemia–reperfusion [44]. However, conflicting data have been obtained by Huang et al. Inhibition of nitric oxide synthase worsened lung injury when ischemia was performed during normoxic ventilation, while it attenuated lung injury during hypoxic ischemia [45]. Very interesting data have been published by Lefer et al. [46,47] who that low concentrations of a in the reaction of nitric oxide with leukocyte–endothelial cell interactions of concentrations of to postischemic hearts attenuated contractile dysfunction, and reduced neutrophil and size induces a microvascular inflammatory response that results in endothelial activation, with induction of a increased of leukocytes and to the vessel and of Reduced and release of mediators by activated endothelium and blood cells result in tissue In addition to the large number of mediators that have been involved in the of this phenomenon, release of carbon monoxide has recently been suggested to play a major role in the hemodynamic alterations of HO-1 expression is stimulated in rat arteries during with an increase in enzyme activity in vascular smooth muscle cells, and inhibition of HO-1 induction also Interestingly, HO-1 induction in vessel walls during seems to be of nitric oxide production is also released in the during and of A may reduce leukocyte recruitment and injury microvascular occurs leukocytes to to and can be by that with Factor or nitric oxide synthase have also been involved in the of as and and in is a major of the diabetic clinical as it is of as and peripheral The of microvascular alterations in diabetes has not been but it is that differences between 1 and 2 and that endothelial dysfunction plays a major role in this phenomenon, with alterations and In addition, since insulin may affect vascular tone and smooth muscle cell insulin resistance might also be involved in this In patients with that are at of it has recently been shown that microvascular response is and it is with insulin suggesting that may have detrimental effects on microvascular function in the and their in where diabetic microvascular injury is more as suggesting that interaction of and may be important in the of diabetic microvascular alterations In diabetic there is a between of retinopathy and reduction of coronary flow with reduction being more when retinopathy is present Thus, of diabetic retinopathy may be a marker of microvascular and reduced coronary flow to evidence is supporting a role for pericytes in diabetic to of pericytes and endothelial cells This phenomenon seems specific for vascular cells and evidence of retinopathy Interestingly, it can also be induced by suggesting a role for blood in this phenomenon Exposure to high blood induces generation of by of and of pericytes to results in of also induces through the induction of and may play a role in development and of diabetic is another associated with microvascular alterations reported in reduced and of and increased vascular all these have been involved in the of and when with a high In these animals, coronary microvessels exhibit a specific of endothelial since vasodilation may still be present when response to and is A similar of vasodilation has also been in patients In the microvasculature of development of is accompanied by and plasma concentrations with mean arterial pressure These also exhibit vasodilation after high This phenomenon seems to be to reduced cGMP and results in and development of vascular resistance in occurs mostly in microvessels, and studies have investigated control of microvascular growth and function is involved in the of has been in the of In subjected to or to reduced vessel with a distribution In patients with essential hypertension, with high blood pressure also have high blood pressure have vasodilation in the and capillaries on the of Recent evidence has also suggested that mediators of may microvascular and growth may also be involved in of arterial pressure, since in their administration increases arterial and blood flow in microvessels and in arterioles of via nitric mechanisms effects on microvasculature are also by which has been involved in in the of of 1 induces and while 2 and induce vasodilation is also accompanied by microvascular which reduction of and smooth muscle cells, and increase in of the vessel morphology has been in brain microvessels A reduced ability to regulate blood flow has also been although has been focused on microcirculation has also been associated with of endothelial which can be by administration of In and microvessels increased release of soluble protein in with control may also be affected during Reduced vasodilation has been reported in microvessels to oxidized response was by with oxygen radical suggesting a role for stress in this phenomenon microvascular function in and these changes can be by These data are with studies showing that induced vascular alterations be by The mechanisms by which venous induces is still in the against an of diffusion of oxygen and while it has been hypothesized that leukocyte activation may be with endothelial injury and tissue inflammation In affected microvascular changes have been with endothelial activation and may induce of more activated of neutrophils and in circulation, that would in turn result in tissue injury expression of which adhesion of neutrophils to vascular can be found in the venous while it is in the In patients with venous causes ischemia and reperfusion, and it has thus been hypothesized that of ischemia–reperfusion may induce an inflammatory reaction, with leukocyte activation and tissue and of venous However, does not release of proinflammatory mediators, suggesting that ischemia–reperfusion not contribute to inflammatory injury in these patients In the 1996 issue devoted to microcirculation there was no review on the role of microcirculation in angiogenesis. However, the mechanisms of and role in a number of pathophysiological conditions have received increasing by the number of papers that were from the for the years articles also involved microcirculation. focused on during tumor and have not been in the present the role of in diabetes and has been between and endothelium during angiogenesis. is an enzyme that migration of endothelial cells during angiogenesis. The form of accumulates in the of microvascular endothelial cells, but not in where it accumulates in from which it may be released In response to microvascular endothelial cells expression of and form while endothelium does not stress occurs under various conditions associated with angiogenesis, and it has been shown that it may angiogenesis. Exposure to induces of cultured endothelium in This phenomenon is mediated by activation of the factor which in turn and of and stimulates This interesting observation may account for induction in various pathophysiological conditions. Taken together, studies on microcirculation have a contribution to of and to a better understanding of a variety of It is that the interest increase further in the