Communication between endothelial and smooth muscle cells

Vascular endothelial cells play a fundamental role in the control of vascular tone, and therefore in the control of local blood flow, by releasing various contracting (endothelin, prostaglandins) and relaxing (prostacycline, NO) factors. An additional mechanism involving the hyperpolarization of the vascular smooth muscle cells is observed mainly in the coronary vascular bed and in the periphery. This phenomenon was attributed to an elusive endothelial factor called endothelium-derived hyperpolarizing factor (EDHF). This mechanism is now better understood. It involves first an increase in the endothelial intracellular concentration of calcium, the activation of endothelial potassium channels and the resulting hyperpolarization of the endothelial cells. The hyperpolarization of the endothelial cells is transmitted to the smooth muscle cells by different pathways. This hyperpolarization propagates along the vessels not only via the smooth muscle cells but also via the endothelial cells. Therefore, the endothelial layer can also be considered as a conducting tissue. The discovery of specific inhibitors of the endothelial cell hyperpolarization allows the assessment of the contribution of EDHF-mediated responses in the control of vascular tone.     

The mitogen-activated protein kinase pathway contributes to vanadate toxicity in vascular smooth muscle cells

Vanadate has been considered in the treatment of diabetes because of its insulin-like effects. However, it has severe toxic effects in both animal and man. In cultured cells, vanadate can either cause death or be growth stimulatory, depending on the cell type and growth conditions. Here, we report that in baboon aortic smooth muscle cells (SMCs), vanadate induced p42/p44 mitogen-activated protein kinase (MAPK) activity. This effect was abolished in the presence of the specific MAPK kinase (MAPKK) inhibitor PD098059. Although activation of p42/p44^MAPK/MAPKK is generally thought to be necessary for proliferation, in SMCs, vanadate did not promote DNA synthesis and inhibited thymidine incorporation stimulated by platelet-derived growth factor (PDGF)-BB in a dose dependent fashion (IC₅₀: 30 M). Prolonged exposure to vanadate exerted cytotoxic effects. Cells retracted, rounded up and detached from the substratum. These vanadate-induced morphological changes were blocked in the presence of PD098059. The addition of PDGF-BB further activated p42/p44^MAPK/MAPKK in the presence of vanadate and substantially increased vanadate toxicity. We conclude from these observations that activation of the p42/p44^MAPK/MAPKK signalling module contributes to the cytotoxic effects induced by vanadate.     

Enzymatic responses to radiation in cultured vascular endothelial and smooth muscle cells

Confluent monolayers of bovine aortic endothelial and smooth muscle cells were exposed to 0-5.0 Gy of 60Co gamma rays. From 0 to 72 hr after irradiation, the monolayer and culture medium were analyzed for cell (nuclei) number, DNA and protein content, the activities of angiotensin converting enzyme (ACE), lactate dehydrogenase (LDH), and superoxide dismutase (SOD), and LDH isoenzyme profile. Irradiated endothelial cells exhibited a time- and dose-dependent increase in cell detachment, decreased DNA and protein content and reduced ACE active per attached cell, increased LDH and SOD activities per microgram of DNA, and increased LDH activity in the culture medium. The latter was accompanied by a shift from LDH 1 to LDH 4 and 5. The release of LDH activity, observed after 0.5 Gy, was the most sensitive endothelial response, and occurred independent of or preceding cell detachment. Vascular smooth muscle cells contained two to three times more SOD activity than did endothelial cells and exhibited no significant responses to 5.0 Gy.     

Papaverine induces apoptosis in vascular endothelial and smooth muscle cells

Papaverine is a vasodilator commonly used in the treatment of vasospasmic diseases such as cerebral spasm associated with subarachnoid hemorrhage, and in the prevention of spasm of coronary artery bypass graft by intraluminal and/or extraluminal administration. In this study, we examined whether papaverine in the range of concentrations used clinically causes apoptosis of vascular endothelial and smooth muscle cells. Apoptotic cells were identified by morphological changes and terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay. In porcine coronary endothelial cells (EC) and rat aortic smooth muscle cells (SMC), papaverine at the concentration of 10(-3) M induced membrane blebbing within 1 hour of incubation. Nuclear condensation and fragmentation were found after 24 hours of treatment. The number of apoptotic cells stained with the TUNEL method was significantly higher in the EC and the SMC after 24 hours of incubation with papaverine at the concentrations of 10(-4) and 10(-3) M than their respective controls. Acidified saline solution (pH 4.8, as control for 10(-3) M papaverine hydrochloride) did not cause apoptosis in these cells. These results showed that papaverine could damage endothelial and smooth muscle cells by inducing changes which are associated with events leading to apoptosis. Since integrity of endothelial cells is critical for normal vascular function, vascular administration of papaverine for clinical use, especially at high concentrations (> or = 10(-4) M), should be re-considered.     

Small GTP binding protein Ras contributes to norepinephrine-induced mitogenesis of vascular smooth muscle cells

Norepinephrine stimulates release of arachidonic acid from tissue lipids. Arachidonic acid metabolites generated through the lipoxygenase and cytochrome P-450 pathways but not cyclooxygenase stimulate mitogen activated protein (MAP) kinase activity and proliferation of vascular smooth muscle cells (VSMC). Moreover, norepinephrine has been shown to activate the Ras/MAP kinase pathway through generation of cytochrome P450 metabolite of arachidonic acid, 20-hydroxyeicosatetraenoic acid (20-HETE). The purpose of this study was to investigate the contribution of Ras in norepinephrine-induced mitogenesis in aortic VSMC. Farnesylation of Ras by farnesyl transferase is required for its full activation. Norepinephrine-induced DNA synthesis, as measured by [3H]-thymidine incorporation, was attenuated by inhibitors of Ras farnesyl transferase FPT III and BMS-191563. These agents also inhibited 20-HETE-stimulated [3H]-thymidine incorporation. In cells transiently transfected with dominant negative Ras (RasN17), norepinephrine, and 20-HETE-induced proliferation of VSMC was attenuated. Both norepinephrine and 20-HETE increased localization of Ras to plasma membrane and MAP kinase activity; FPT III attenuated these effects. These data suggest that VSMC proliferation induced by norepinephrine and 20-HETE is mediated by Ras/MAP kinase pathway.     

The extracellular matrix and the control of proliferation of vascular endothelial and vascular smooth muscle cells

In this short review we describe the observations which have led us to conclude that one of the most important components involved in modulating cell proliferation in vitro, and probably in vivo as well, may be the extrac-cellular matrix upon which cells rest.     

Co-cultivation of vascular endothelial and smooth muscle cells using microcarrier techniques

A system is described which uses microcarrier culture techniques for the co-cultivation of different cell types without direct contact between cell populations. In co-cultivation, arterial endothelial cells induced proliferation in > 90% of quiescent homologous arterial smooth muscle cells in the absence of serum-derived growth factors. The microcarrier coculture system allows investigation of potent local humoral interactions between vascular cells in vitro.     

Heparin interactions with cultured human vascular endothelial and smooth muscle cells: incidence on vascular smooth muscle cell proliferation

The binding, internalization, and metabolism of [3H]-heparin by human umbilical vein endothelial cells (HUVEC) and human umbilical arterial smooth muscle cells (HUASMC) have been characterized using size-exclusion HPLC. Incubation of HUVEC with [3H]-heparin demonstrated selective binding of high-molecular-weight (MW) components (MW = 21 kd), which was followed by rapid, temperature-dependent internalization. Over the next 3 hours, this internalized [3H]-heparin was degraded to low-MW fragments (MW = 0.9 kd). Primary cultures of HUASMC selectively bound extremely high-MW components (MW = 40 kd) and also smaller components whose MW (0.9 kd) corresponded to that of the heparin metabolite(s) formed by HUVEC. Subcultured HUASMC bound only the 40-kd components. Internalization of heparin by smooth muscle cells (SMC) was significantly slower than that determined for HUVEC, and even after 4 hours there was no evidence of the heparin being metabolized. However, when incubating primary rabbit aortic SMC with purified low-MW heparin fragment(s) produced in culture by HUVEC, a significantly lower proliferative response of these cells (IC50 = 18.4 micrograms/ml) was obtained. Virtually no effect was observed with subcultured SMC in the range of the tested concentrations (0-20 micrograms/ml). These fragments were 10- to 15-fold more effective in inhibiting primary SMC growth than was standard heparin. Furthermore, heparin fractions in the same range of molecular weights, purified either after nitrous acid or heparinase depolymerization of standard heparin, showed no activity on primary SMC growth, thus indicating a high degree of selectivity of the heparin metabolite(s) produced by HUVEC in culture.     

Carnitine requirement of vascular endothelial and smooth muscle cells in imminent ischemia

Rats were subjected to bilateral carotid artery occlusion for 30 min, followed by reperfusion for varying time periods. The concentration of reduced and oxidized glutathione, glutathione peroxidase and glutathione reductase were determined in whole brain after varying periods of reperfusion. Lipid peroxidation was also assessed by determining the levels of malondialdehyde (MDA) in the brain. Reperfusion for 1 hr following bilateral carotid artery occlusion resulted in significant decrease in total glutathione (GSH) concentration along with small but significant increase in oxidized glutathione (GSSG) levels. After 4 hr of reperfusion, GSH levels recovered, although GSSG levels remained elevated up to 12 hr of reperfusion. Increase in malondialdehyde levels was also detected in the brain up to 12 hr of reperfusion. Glutathione reductase activity remained significantly low up to 144 hr of reperfusion, while glutathione peroxidase activity remained unaffected. These results demonstrate that oxidative stress is generated in the brain during reperfusion following partial ischemia due to bilateral carotid artery occlusion.     

Upregulation of connexin43 gap junctions between neointimal smooth muscle cells

Increased expression of connexin43 gap junctions in smooth muscle cells (SMC) is implicated in the response to primary arterial injury and in the early stages of human coronary atherosclerosis, but the relevance of these findings to restenosis is unknown. Here we investigated the expression of connexin43 gap junctions in restenotic aortas of cholesterol-fed double injured rabbits. Immunofluorescence confocal microscopy was used to evaluate temporal and spatial expression patterns and to characterize the major expressing cell type. Parallel studies were conducted by electron microscopy, in situ hybridization and Northern blot analysis. Connexin43 gap junctions- and connexin43 mRNA-expressing cells were abundant in the media of non-injured control aorta. Following primary injury and 6 weeks cholesterol diet, connexin43 gap junctions were found distributed throughout the primary intimal layer; although medial expression was reduced, the overall mRNA expression level remained similar to that of non-injured controls. After secondary injury, no major change in distribution pattern of connexin43 gap junctions occurred up to day 10, when marked neointimal labeling was observed. This overall pattern persisted, though with some diminution, at later stages. On the mRNA level total connexin43 mRNA expression declined to about 40% of control values within 4 days after secondary injury (P < 0.05), but subsequently increased four-fold, attaining levels double that of non-injured controls in the 10-day group (P < 0.005 versus control and 4 days). At later stages mRNA expression levels returned to values similar to those of non-injured controls. At all stages, connexin43 gap junctions were localized to the SMC, not to macrophages. We conclude that the enhanced gap junction formation may contribute to the coordination of the response of SMC after secondary injury, particularly in the early phase of restenosis.     

Lipid transfer between endothelial and smooth muscle cells in coculture

A coculture system was employed to study the interactions between endothelium and vascular smooth muscle cells in arachidonic acid metabolism. Bovine aortic endothelial cells grown on micropore filters impregnated with gelatin and coated with fibronectin are mounted on polystyrene chambers and suspended over confluent smooth muscle cultures. The endothelial basal laminae are oriented toward the underlying smooth muscle, and the two layers are separated by only 1 mm. Each cell layer was assayed individually: apical and basolateral fluid also was collected separately for assay. Fatty acids, including arachidonic acid, are readily transferred between the endothelial and smooth muscle cells in this system. Distribution of the incorporated fatty acids among the lipids of each cell is the same as when the fatty acid is added directly to the culture medium. Arachidonic acid released from endothelial cells is available as a substrate for prostaglandin production by smooth muscle. In addition, fatty acids released from the smooth muscle cells can pass through the endothelium and accumulate in the fluid bathing the endothelial apical surface. These fatty acid interchanges may be involved in cell-cell signaling within the vascular wall, the clearance of lipids from the vascular wall, or the redistribution of arachidonic acid and other polyunsaturated fatty acids between adjacent cell types. Furthermore, the findings suggest that prostaglandin production by smooth muscle cells can occur in response to stimuli that cause arachidonic acid release from endothelial cells.     

Improved arterial wall model by coculturing vascular endothelial and smooth muscle cells

We have constructed an in vitro arterial wall model by coculturing bovine arterial endothelial cells (ECs) and smooth muscle cells (SMCs). When ECs were seeded directly over SMCs and cocultured in an ordinary culture medium, ECs grew sparsely and did not form a confluent monolayer. Addition of ascorbic acid to the culture medium at concentrations greater than 50 μg/ml increased the production of type IV collagen by the SMCs, and ECs formed a confluent monolayer covering the entire surface of SMCs. Histological studies showed that the thickness of the cell layer composed of ECs and SMCs increased with increasing duration of coculture. This arterial wall model, prepared by our method, may serve as a simple and good in vitro model to study the effects of factors such as biological chemicals and shear stress on cell proliferation and other physiological functions of arterial walls.     

Biotransformation of organic nitrates to nitric oxide by vascular smooth muscle and endothelial cells.

The vasodilator action of organic nitrates is thought to be mediated by an increase in the level of cGMP following stimulation of the cytosolic enzyme guanylate cyclase in the vascular smooth muscle cell. However, direct evidence for the formation of the putative active metabolite, nitric oxide (NO) within the different compartments of the vascular wall is still missing. We here demonstrate for the first time that cultured vascular smooth muscle cells as well as endothelial cells from different species actively metabolize organic nitrates to NO. We furthermore present evidence for an outward transport of cGMP from both cell types following stimulation of soluble guanylate cyclase. The rate of NO release closely correlated with the rate of cGMP egression. Biotransformation of organic nitrates to NO appeared to comprise at least two different components, a heat-sensitive enzymatic pathway which is short-lived and prone to rapid desensitization and a second non-enzymatic component which is apparently unsaturable and longer lasting. The marked decrease in the release of NO and cGMP upon the repeated administration of organic nitrates suggests that the phenomenon of "nitrate tolerance" is mainly due to an impaired biotransformation. We propose that the metabolism of nitrates to NO may have important implications for the prevention of atherosclerosis and the therapeutic modulation of blood cell function.     

Induction of indoleamine 2,3-dioxygenase in vascular smooth muscle cells by interferon-gamma contributes to medial immunoprivilege

Atherosclerosis and graft arteriosclerosis are characterized by leukocytic infiltration of the vessel wall that spares the media. The mechanism(s) for medial immunoprivilege is unknown. In a chimeric humanized mouse model of allograft rejection, medial immunoprivilege was associated with expression of IDO by vascular smooth muscle cells (VSMCs) of rejecting human coronary artery grafts. Inhibition of IDO by 1-methyl-tryptophan (1-MT) increased medial infiltration by allogeneic T cells and increased VSMC loss. IFN-gamma-induced IDO expression and activity in cultured human VSMCs was considerably greater than in endothelial cells (ECs) or T cells. IFN-gamma-treated VSMCs, but not untreated VSMCs nor ECs with or without IFN-gamma pretreatment, inhibited memory Th cell alloresponses across a semipermeable membrane in vitro. This effect was reversed by 1-MT treatment or tryptophan supplementation and replicated by the absence of tryptophan, but not by addition of tryptophan metabolites. However, IFN-gamma-treated VSMCs did not activate allogeneic memory Th cells, even after addition of 1-MT or tryptophan. Our work extends the concept of medial immunoprivilege to include immune regulation, establishes the compartmentalization of immune responses within the vessel wall due to distinct microenvironments, and demonstrates a duality of stimulatory EC signals versus inhibitory VSMC signals to artery-infiltrating T cells that may contribute to the chronicity of arteriosclerotic diseases.     

Adenylyl cyclase isoforms and signal integration in models of vascular smooth muscle cells

Adenylyl cyclases present a potential focal point for signal integration in vascular smooth muscle cells (VSMC) influencing contractile state and cellular responses to vessel wall injury. In the present study, we examined the influence of the vasoactive peptide arginine vasopressin (AVP) on cAMP regulation in primary cultures of rat aortic VSMC and in the A7r5 arterial smooth muscle cell line. In cultured VSMC and A7r5 cells, AVP had no effect on basal cAMP but differentially affected beta-adrenergic receptor-induced activation of adenylyl cyclase. AVP synergistically increased (twofold) isoproterenol-stimulated cAMP production in VSMC but inhibited the effect of isoproterenol (50%) in the A7r5 cell line. The effects of AVP in both preparations were blocked when cells were pretreated with a selective V(1) vasopressin receptor antagonist. Moreover, the actions of AVP in both models were dependent on release of intracellular Ca(2+) and were mimicked by elevation of Ca(2+) with the ionophore A23187, suggesting that the responses to AVP involve Ca(2+)-mediated regulation of adenylyl cyclase stimulation. Adenylyl cyclase types I, III, and VIII are stimulated by Ca(2+)/calmodulin, whereas types V and VI are directly inhibited by Ca(2+). RNA blot analysis for effector isotypes indicated that both VSMC and A7r5 cells expressed types III, V, and VI. VSMC also expressed mRNA for type IV and VIII effectors, which could account for the cell-specific responses to peptide hormone and Ca(2+).     

A model for studying the effect of shear stress on interactions between vascular endothelial cells and smooth muscle cells

Vascular endothelial cells (ECs) are constantly subjected to blood flow-induced shear stress and the influences of neighboring smooth muscle cells (SMCs). In the present study, a coculture flow system was developed to study the effect of shear stress on EC-SMC interactions. ECs and SMCs were separated by a porous membrane with only the EC side subjected to the flow condition. When ECs were exposed to a shear stress of 12 dynes/cm2 for 24 h, the cocultured SMCs tended to orient perpendicularly to the flow direction. This perpendicular orientation of the cocultured SMCs to flow direction was not observed when ECs were exposed to a shear stress of 2 dynes/cm2. Under the static condition, long and parallel actin bundles were observed in the central regions of the cocultured SMCs, whereas the actin filaments localized mainly at the periphery of the cocultured ECs. After 24 h of flow application, the cocultured ECs displayed very long, well-organized, parallel actin stress fibers aligned with the flow direction in the central regions of the cells. Immunostaining of platelet endothelial cell adhesion molecule-1 confirmed the elongation and alignment of the cocultured ECs with the flow direction. Coculture with SMCs under static condition induced EC gene expressions of growth-related oncogene-alpha and monocyte chemotactic protein-1, and shear stress was found to abolish these SMC-induced gene expressions. Our results suggest that shear stress may serve as a down-regulator for the pathophysiologically relevant gene expression in ECs cocultured with SMCs.     

Metabolic cooperation between vascular endothelial cells and smooth muscle cells in co-culture: changes in low density lipoprotein metabolism

A microcarrier co-culture system for aortic endothelial cells and smooth muscle cells (SMCs) was developed as a model for metabolic interactions between cells of the vessel wall. Low density lipoprotein (LDL) metabolism in SMCs was significantly influenced by co-culture with endothelium. The numbers of high affinity receptors for LDL was increased more than twofold (range, 2.1-5.6), with concomitant increases in LDL receptor-mediated endocytosis and degradation. These effects reached a plateau at an endothelial cell/SMC ratio of 1. Kinetic analysis of the endocytic pathway for LDL in SMCs indicated that, in co-culture with endothelium, there was no alteration in the binding affinity of LDL to its receptors but that the internalization rate constant declined and the rate constant for degradation increased. This analysis suggested that the formation and migration of endocytic vesicles was the rate-limiting step of enhanced LDL metabolism under co-culture conditions. Two mechanisms by which endothelial cells influenced smooth muscle LDL metabolism were identified. First, mitogen(s) derived from endothelial cells stimulated entry of SMCs into the growth cycle, and the changes in LDL metabolism occurred as a consequence of G1-S transition. Second, SMC lipoprotein metabolism was stimulated in the absence of mitogens by a low molecular weight (less than 3,500) factor or factors. Co-culture was a required condition for the latter effect, suggesting that the mediator(s) may be unstable or that cell-cell communication was necessary for expression. These results (a) demonstrate that vascular cell interactions can modify LDL metabolism in SMCs, (b) provide some insights into the mechanisms responsible, and (c) identify co-culture as an experimental approach appropriate to certain aspects of vascular cell biology.