T cell-associated α4ß7 but not α4ß1 integrin is required for the induction and perpetuation of chronic colitis.

Anti-adhesion therapies that target α(4) integrins (e.g., natalizumab) are thought to work by blocking T-cell recruitment to the intestinal tissues in patients with Crohn's disease (CD); however, little direct evidence is available to confirm this contention. We wished to evaluate the importance of T cell-associated α(4) integrins in a chronic colitis model in mice and to determine the effect of natalizumab treatment on intestinal tissue T-cell accumulation in human CD. Adoptive transfer of T cells lacking α(4) (α(4)(-/-)) but not ß(1) integrin into immunodeficient mice produced significantly attenuated disease. This was correlated with reduced numbers of colon CD4 T cells compared with the control mice; however, tissue distribution of T helper type 1 (Th1) and T helper type 17 (Th17) cells and regulatory T cells (Tregs) was not affected by the lack of α(4). Furthermore, α(4)(-/-) T cells demonstrated defective homing to the chronically inflamed small intestines and colons. Finally, patients treated with natalizumab showed significant reduction in mucosal CD4 T cells and no skewing in the foxp3(+) Treg or T-bet(+)Th1 fractions thereof. These results demonstrate a direct role for T cell-associated α(4)ß(7) but not α(4)ß(1) integrins during initiation and perpetuation of chronic colitis. Moreover, our data demonstrated that natalizumab treatment reduced mucosal CD4 T-cell accumulation in CD patients.

Deficiency of CD18 gene reduces brain edema in experimental intracerebral hemorrhage in mice.

Experimental studies of intracerebral hemorrhage (ICH) point toward leukocytes as a major contributor to ICH-induced brain injury. Leukocyte and endothelial cell adhesion molecules are responsible for injurious neutrophil-endothelial cell interactions in vasculature. Since deficiency of leukocyte-expressed CD18 protects against ischemia-reperfusion injury, we hypothesized that such deficiency may have similar effect in ICH-induced injury. Our aim was to investigate whether CD18 deficiency affords neuroprotection by decreasing ICH-induced brain injury, thereby improving neurological function and reducing mortality. A total of 20 males wild-type CDI8+/+ mice and 12 CD18-/- knockout mice were used in our study. ICH was induced by collagenase injection. Mortality, neurological function, and brain edema were measured at 24h after ICH. Data were analyzed by ANOVA, Chi-square, and Student t-test. Differences of p < 0.05 were considered statistically significant. Our study showed that the increase in brain water content caused by ICH was significantly smaller in CD18 knockout mice compared with wild-type mice (p < 0.05, Student t-test). This result correlated with a tendency toward improvement of neurological function and a decrease in mortality. We conclude that CD18 deficiency significantly reduces brain edema after ICH, which corresponds with a trend toward reduction in neurological deficit and mortality.

Regulation of endothelial glutathione by ICAM-1: implications for inflammation.

The role of glutathione (GSH) in inflammation is largely discussed from the context of providing reducing equivalents to detoxify reactive oxygen and nitrogen species. Inflammation is now recognized to be an underlying cause of many vascular diseases including atherosclerosis, a disease in which endothelial GSH concentrations are decreased. However, mechanisms that control GSH levels are poorly understood. Key players in the inflammatory process are endothelial adhesion molecules, including intercellular adhesion molecule-1 (ICAM-1). This adhesion molecule is present constitutively and can be induced by a variety of inflammatory stimuli. In this study, using mouse aortic endothelial cells (MAEC) deficient in ICAM-1, we demonstrate a novel interplay between constitutive ICAM-1 and cellular GSH. Deficiency of ICAM-1 was associated with an approximately twofold increase in total GSH content. Inhibiting glutamate-cysteine ligase (GCL), the enzyme that catalyses the rate-limiting step in GSH biosynthesis, prevented the increase in GSH. In addition, the catalytic subunit of GCL was increased (approximately 1.6-fold) in ICAM-1 deficient relative to wild-type cells, suggesting that constitutive ICAM-1 represses GCL expression. Furthermore, the ratio of reduced (GSH) to oxidized (GSSG) glutathione was also increased suggesting a role for ICAM-1 in modulating cellular redox status. Interestingly, increasing cytosolic GSH in wild-type mouse endothelial cells decreased constitutive ICAM-1, suggesting the presence of an inverse and reciprocal pathway. To test the effects of inducible ICAM-1 on GSH, cells were stimulated with the proinflammatory cytokine TNF-alpha. TNF-alpha stimulated production of ICAM-1, which was however not associated with induction of GSH. In contrast, supplementation of endothelial cells with GSH before TNF-alpha addition, inhibited induction of ICAM-1. These data suggest a novel regulatory pathway between constitutive ICAM-1 and GSH synthesis in the endothelium and are discussed in the context of modulating the inflammatory response.

H(2)O(2)-mediated permeability II: importance of tyrosine phosphatase and kinase activity.

We previously reported that exposure of endothelial cells to H(2)O(2) results in a loss of cell-cell apposition and increased endothelial solute permeability. The purpose of this study was to determine how tyrosine phosphorylation and tyrosine phosphatases contribute to oxidant-mediated disorganization of endothelial cell junctions. We found that H(2)O(2) caused a rapid decrease in total cellular phosphatase activity that facilitates a compensatory increase in cellular phosphotyrosine residues. H(2)O(2) exposure also results in increased endothelial monolayer permeability, which was attenuated by pp60, an inhibitor of src kinase. Inhibition of protein tyrosine phosphatase activity by phenylarsine oxide (PAO) demonstrated a similar permeability profile compared with H(2)O(2), suggesting that tyrosine phosphatase activity is important in maintaining a normal endothelial solute barrier. Immunofluorescence shows that H(2)O(2) exposure caused a loss of pan-reactive cadherin and beta-catenin from cell junctions that was not blocked by the src kinase inhibitor PP1. H(2)O(2) also caused beta-catenin to dissociate from the endothelial cytoskeleton, which was not prevented by PP1. Finally, we determined that PP1 did not prevent cadherin internalization. These data suggest that oxidants like H(2)O(2) produce biological effects through protein phosphotyrosine modifications by decreasing total cellular phosphatase activity combined with increased src kinase activity, resulting in increased endothelial solute permeability.

Essential role of ICAM-1 in mediating monocyte adhesion to aortic endothelial cells.

Monocyte-endothelial cell interactions have been implicated in the pathogenesis of a number of vascular diseases that target arterial and aortic endothelium, including atherosclerosis. Many different adhesion molecules, such as intercellular adhesion molecule (ICAM)-1, are thought to mediate monocyte binding to endothelial cells during the development of these diseases. However, conflicting results have been reported regarding the specific role of ICAM-1 in these events. In this study, we used a genetic approach to determine the contribution of ICAM-1 in mediating monocyte adhesion to mouse aortic endothelial cells (MAEC) derived from both wild-type and ICAM-1(-/-) mice. Treatment of wild-type MAEC with oxidized low-density lipoprotein significantly induced both WEHI 274.1 and whole blood monocyte adhesion, whereas similarly treated ICAM-1(-/-) MAEC showed a complete inhibition of monocyte binding. Dose-response treatment with tumor necrosis factor-alpha also increased monocyte adhesion to wild-type MAEC, but significant adhesion was only observed at higher doses for ICAM-1(-/-) MAEC. These data demonstrate a crucial role for ICAM-1-mediated monocyte-endothelial cell interactions in response to specific stimuli involved in inflammatory vascular diseases.

In vitro culture and characterization of gene targeted mouse endothelium.

Endothelial cells play a crucial role in maintaining cardiovascular homeostasis. Although many cardiovascular disorders involve endothelial cell dysfunction, the specific cellular and molecular mechanisms involved are not well known. We sought to establish a reproducible method of endothelial cell isolation from gene targeted mice to specifically examine endothelial pathophysiological mechanisms. Primary aortic endothelial cell cultures were established from wild type and intercellular adhesion molecule-1 (ICAM-1) deficient mice. Isolation of mouse aortic endothelial cells (MAEC) by fluorescent activated cell sorting routinely resulted in pure, homogenous, primary cultures. Wild type and ICAM-1 deficient endothelial cell morphology was similar, with both cultures showing cobblestone morphology and DiI-Ac-LDL staining. Monocyte adhesion to ICAM-1 deficient aortic endothelial cells was decreased by 86% as compared with wild type MAEC. Monocyte adhesion was also determined using YN-1, an ICAM-1 blocking antibody. YN-1 decreased monocyte adhesion to wild type aortic endothelial cells by 25%, whereas YN-1 did not further decrease monocyte adhesion to ICAM-1 deficient MAEC. These data demonstrate that gene targeted endothelial cell cultures are an effective means of identifying specific cellular and molecular mechanisms involved in endothelial cell physiology and dysfunction.

The role of p38 MAP kinase in hydrogen peroxide mediated endothelial solute permeability.

OBJECTIVE: The purpose of this study was to determine the contribution of p38 MAP kinase activity during hydrogen peroxide mediated increased endothelial solute permeability. We also sought to identify the role of p38 MAP kinase-mediated changes in endothelial cell architecture due to hydrogen peroxide challenge. METHODS: Hydrogen peroxide mediated permeability of HUVEC was determined with and without inhibition of p38 MAP kinase by SB202190. Hydrogen peroxide mediated rearrangement of the endothelial actin cytoskeleton and junctional proteins occludin and ZO-1 were observed by immunofluorescence microscopy. RESULTS: Hydrogen peroxide treatment of endothelial monolayers caused a significant increase in solute permeability over a ninety-minute time period. Oxidant-mediated permeability and phosphorylation of p38 MAP kinase was significantly attenuated by SB 202190. Immunofluorescent staining for the tight junctional proteins occludin and ZO-1 demonstrated that oxidant challenge caused a loss of endothelial tight junction organization. Rhodamine phalloidin staining of the actin cytoskeleton showed that hydrogen peroxide stimulated increased stress fiber formation with concomitant gap formation between adjacent endothelial cells. Inhibition of p38 MAP kinase during oxidant challenge significantly attenuated actin stress fiber formation and prevented gap formation. CONCLUSIONS: These data demonstrate that p38 MAP kinase activity is involved in hydrogen peroxide mediated permeability, stress fiber formation, and intracellular gap formation.

Leukocyte and endothelial cell adhesion molecules in a chronic murine model of myocardial reperfusion injury.

Expression of endothelial and leukocyte cell adhesion molecules is a principal determinant of polymorphonuclear neutrophil (PMN) recruitment during inflammation. It has been demonstrated that pharmacological inhibition of these molecules can attenuate PMN influx and subsequent tissue injury. We determined the temporal expression of alpha-granule membrane protein-40 (P-selectin), endothelial leukocyte adhesion molecule 1 (E-selectin), and intercellular cell adhesion molecule 1 (ICAM-1) after coronary artery occlusion and up to 3 days of reperfusion. The expression of all of these cell adhesion molecules peaked around 24 h of reperfusion. We determined the extent to which these molecules contribute to PMN infiltration by utilizing mice deficient (-/-) in P-selectin, E-selectin, ICAM-1, and CD18. Each group underwent 30 min of in vivo, regional, left anterior descending (LAD) coronary artery ischemia and 24 h of reperfusion. PMN accumulation in the ischemic-reperfused (I/R) zone was assessed using histological techniques. Deficiencies of P-selectin, E-selectin, ICAM-1, or CD18 resulted in significant (P < 0.05) attenuation of PMN infiltration into the I/R myocardium (MI/R). In addition, P-selectin, E-selectin, ICAM-1, and CD18 -/- mice exhibited significantly (P < 0.05) smaller areas of necrosis after MI/R compared with wild-type mice. These data demonstrate that MI/R induces coronary vascular expression of P-selectin, E-selectin, and ICAM-1 in mice. Furthermore, genetic deficiency of P-selectin, E-selectin, ICAM-1, or CD18 attenuates PMN sequestration and myocardial injury after in vivo MI/R. We conclude that P-selectin, E-selectin, ICAM-1, and CD18 are involved in the pathogenesis of MI/R injury in mice.

H(2)O(2)-mediated permeability: role of MAPK and occludin.

H(2)O(2)-mediated elevation in endothelial solute permeability is associated with pathological events such as ischemia-reperfusion and inflammation. To understand how H(2)O(2) mediates increased permeability, we investigated the effects of H(2)O(2) administration on vascular endothelial barrier properties and tight junction organization and function. We report that H(2)O(2) exposure caused an increase in endothelial solute permeability in a time-dependent manner through extracellularly regulated kinase 1 and 2 (ERK1/ERK2) signal pathways. H(2)O(2) exposure caused the tight junctional protein occludin to be rearranged from endothelial cell-cell junctions. Occludin rearrangement involved redistribution of occludin on the cell surface and dissociation of occludin from ZO-1. Occludin also was heavily phosphorylated on serine residues upon H(2)O(2) administration. H(2)O(2) mediates changes in ERK1/ERK2 phosphorylation, increases endothelial solute permeability, and alters occludin localization and phosphorylation were all blocked by PD-98059, a specific mitogen-activated protein (MAP) or ERK kinase 1 inhibitor. These data strongly suggest that H(2)O(2)-mediated increased endothelial solute permeability involves the loss of endothelial tight junction integrity through increased ERK1/ERK2 activation.

Inflammatory mediators induce sequestration of VE-cadherin in cultured human endothelial cells.

The mechanisms through which inflammatory mediators modify endothelial junctional structure are not well understood. Endothelial cells exposed to 1 mM H2O2, 0.1 mM histamine or 4 mM EDTA displayed decreased amounts of VE-cadherin on the cell surface in a time-dependent manner. H2O2 and EDTA-treated cells showed a sustained reduction in surface VE-cadherin, but histamine (0.1 mM) decreased cell surface VE-cadherin only at 5 and 15 min, not at 30 and 60 min. Sequestering of VE-cadherin could also be visualized as a decrease in immunofluorescent labeling of endothelial junctions in fixed, non-extracted monolayers. However, junctional staining was observed in these cells after membrane extraction. This decreased surface expression of VE-cadherin was actin-filament, but not PKC/MAP kinase dependent. VE-cadherin binding to the cytoskeleton was decreased by EDTA, but was not diminished by histamine or H2O2. Therefore, by promoting sequestration of junctional cadherins, inflammatory mediators may decrease adhesive bonds between apposed endothelial cells and increase solute permeability.

Effect of reactive oxygen metabolites on endothelial permeability: role of nitric oxide and iron.

OBJECTIVE: We evaluated the effects of the xanthine oxidase (XO)-derived reactive oxygen metabolites on the permeability of bovine pulmonary artery-endothelial monolayers and examined how iron and nitric oxide (NO) participate in these changes in permeability. METHODS: Permeability was measured using a cell-column chromatographic method in which monolayers were exposed to combinations of agents. RESULTS: Exposure of monolayers to a superoxide/peroxide generator, xanthine (X, 0.1 mM)/XO (25 mU/mL), increased solute permeability after 10 minutes, but the same dose of either X or XO alone did not. Exposure of monolayers to peroxide (0.1 mM) also increased permeability, but only after 70 minutes. This X/XO permeability was attenuated by either catalase, superoxide dismutase, methionine (1 mM), an oxy-radical scavenger, or desferrioxamine (0.1 mM), an iron chelator. Spermine NONOate (SNO), an NO donor, attenuated X/XO permeability at 0.1 mM, but this protection was not significant at 0.01 or 1 mM. Spermine NONOate (0.1 mM) did not alter the permeability produced by 0.1 mM peroxide. L-N5-(1-iminoethyl)-ornithine (10 microM), an NO synthase inhibitor, completely blocked peroxide-, and partially attenuated X/XO-mediated permeability. However, 3-morphosynodiomine (SIN-1, 1 mM) plus catalase (1,000 U/mL), a peroxynitrite generator, did not alter permeability. CONCLUSIONS: Xanthine/Xanthine Oxidase permeability involves peroxide, superoxide, oxy-radicals, and iron. Endogenous NO may regulate peroxide-, but not superoxide-mediated permeability. The protective effects of exogenous NO on the X/XO permeability may represent interactions between superoxide, peroxide, and cell surface-bound iron.

Organ preservation solutions increase endothelial permeability and promote loss of junctional proteins.

OBJECTIVE: To investigate the effects of the organ preservation solutions UW and Plegisol on endothelial permeability; occludin and vascular endothelial (VE)-cadherin content in human umbilical vein endothelial cells (HUVEC); and junctional localization of these proteins after exposure to these solutions. SUMMARY BACKGROUND DATA: Organ preservation for transplantation is limited by several challenges, including loss of tissue function, tissue injury, and tissue edema. Occludin and VE-cadherin are responsible for maintaining and regulating the endothelial solute barrier. Several studies have noted organ edema and dysfunction with preservation, as well as gaps between endothelial cells suggesting that disorganization of junctional proteins (e.g., occludin and VE-cadherin) is responsible for interstitial edema. METHODS: HUVEC monolayers were treated with 4 degrees C UW and Plegisol for 3 and 6 hours and then reperfused with normal buffer. Permeability was examined using FITC-dextran tracer during the reperfusion phase. Occludin and VE-cadherin content at different time points was measured by Western blotting. Treated groups were also examined by immunofluorescence for occludin, VE-cadherin, and F-actin. RESULTS: Compared with untreated controls, cold preservation for 3 and 6 hours increased endothelial permeability after rewarming, which appears to depend on the duration of cold exposure. Monolayers exposed to 3 hours of cold preservation did not have increased permeability in the first hour after rewarming but had significantly increased permeability after the first hour and all subsequent time points. Monolayers exposed to 6 hours of cold preservation had increased permeability after the first hour and at all later time points. Western blotting demonstrated that occludin content was decreased to a similar extent with all solutions after 3 hours of cold preservation. Six hours of cold preservation in Plegisol reduced the occludin content significantly compared with UW and control. VE-cadherin content was unchanged after 3 hours of cold preservation but was dramatically reduced in all groups at 6 hours. Immunofluorescent staining demonstrated junctional gap formation and discontinuous staining of occludin and VE-cadherin with all cold preservation protocols; changes in F-actin organization were observed at 3 and 6 hours after cold preservation. CONCLUSION: The changes in occludin, VE-cadherin, and F-actin content and organization and increased permeability associated with cold storage demonstrate that alterations of the tight and adherens junctions may underlie organ edema associated with cold organ preservation. These data also suggest that novel strategies to maintain the content and integrity of endothelial junctional proteins may provide an important therapeutic avenue for organ preservation.

Differential monocyte adhesion and adhesion molecule expression in venous and arterial endothelial cells.

We compared U-937 cell adhesion and adhesion molecule expression in human umbilical venous (HUVECs) and arterial (HUAECs) endothelial cells exposed to tumor necrosis factor (TNF), interleukin-1, and lipopolysaccharide (LPS). TNF and LPS stimulated vascular cell adhesion molecule (VCAM)-1 surface expression and adhesion of U-937 monocyte-like cells to HUVECs but not to HUAECs. Antibody studies demonstrated that in HUVECs at least 75% of the adhesion response is VCAM-1 mediated. Interleukin-1 stimulated U-937 cell adhesion to and VCAM-1 surface expression in both HUVECs and HUAECs. Pyrrolidinedithiocarbamate and the proteasome inhibitor MG-132 blocked TNF- and LPS-stimulated U-937 cell adhesion to HUVECs. These agents also significantly decreased TNF- and LPS-stimulated increases in HUVEC surface VCAM-1. TNF increased VCAM-1 protein and mRNA in HUVECs that was blocked by pyrrolidinedithiocarbamate. However, neither TNF or LPS stimulated VCAM-1 expression in HUAECs. TNF stimulated expression of both intercellular adhesion molecule-1 and E-selectin in HUVECs, but in HUAECs, only intercellular adhesion molecule-1 was increased. Electrophoretic mobility shift assays demonstrated no difference in the pattern of TNF-stimulated nuclear factor-kappaB activation between HUVECs and HUAECs. These studies demonstrate a novel and striking insensitivity of arterial endothelium to the effects of TNF and LPS and indicate a dissociation between the ability of HUAECs to upregulate nuclear factor-kappaB and VCAM-1.

Activated T-lymphocytes express occludin, a component of tight junctions.

T-lymphocytes routinely traffic from the lymphoid and vascular compartments to the tissues during immune surveillance and inflammatory responses. This egress occurs without compromising endothelial barrier, which is maintained by tight junctions (zonula occludens). We report that T-lymphocytes up-regulate the expression of occludin, a major component of the tight junction in response to stimulation with phorbol ester (PMA) + calcium ionophore, CD3 antibody or T-cell receptor (TCR) antibody. Only activated T-lymphocytes express occludin; this adhesion molecule is nearly absent in resting T-lymphocytes. By immunofluorescence, occludin is seen in lymphocyte aggregates, but does not appear to mediate aggregation since only 50% of the cells in these clusters express occludin. Occludin is expressed between 8 and 24 h following stimulation, and persists for at least 48 h. These data indicate that activated T cells produce occludin which may regulate lymphocyte adhesion and trafficking.

Expression of zonula occludens and adherens junctional proteins in human venous and arterial endothelial cells: role of occludin in endothelial solute barriers.

OBJECTIVE: The purpose of this study was to correlate the expression of occludin and VE-cadherin with the solute barrier properties of arterial and venous endothelial monolayers. METHODS: Immunofluorescent confocal and traditional microscopy were used to determine junctional protein localization in endothelium in vivo and in vitro respectively, and western and northern analysis used to determine protein and gene expression levels. Permeability of endothelial monolayers was examined under normal, low calcium, and cytochalasin-D treatment conditions. Antisense oligonucleotide experiments for occludin were performed to determine the contribution of occludin to solute barrier. RESULTS: Occludin protein in endothelial monolayers is more concentrated in arterial junctions than in venous junctions both in vivo and in vitro. Arterial endothelial cells express 18-fold more occludin protein and nine times more occludin mRNA compared to venous endothelial cells. In vivo, both endothelial cells demonstrate VE-cadherin staining; and in vitro, only venous endothelial cells express VE-cadherin protein and mRNA. Occludin antisense experiments suggest that both arterial and venous barrier properties are due to these different amounts of occludin expression. Venous barrier was remarkably sensitive to low extracellular calcium, while arterial barrier was more sensitive to cytochalasin-D. CONCLUSIONS: These findings suggest strongly that arterial and venous endothelial barrier reflects the level of expression of different adhesion molecules and that modulation of these proteins, especially occludin, may regulate the level of endothelial solute barrier.

The role of cadherin endocytosis in endothelial barrier regulation: involvement of protein kinase C and actin-cadherin interactions.

We have previously reported that exposure of endothelial monolayers to low (0.12 mM) extracellular calcium significantly decreased the endothelial solute barrier, and that this effect was reversed by restoring 'normal' (1.2 mM) calcium (1). This effect was shown to be dependent on cadherins, however the molecular mechanisms through which barrier was altered by low calcium were not characterized. Here we investigated the mechanism of increased endothelial permeability produced by low calcium exposure. Endothelial permeability was significantly increased by exposure to low (0.12 mM) calcium; this effect was attenuated by pre-treatment with the protein kinase C (PKC) inhibitor, staurosporine (2 x 10(-7) M) for 30 min. Cell border retraction and gap formation produced by low calcium was also prevented by staurosporine. Treatment of monolayers with 0.12 mM calcium also stimulated the endocytosis of endothelial cadherins. This low calcium mediated cadherin endocytosis was also prevented by pretreatment with staurosporine. Low calcium mediated endocytosis was also prevented by the actin filament toxin, cytochalasin D (1 ug/ml, 30 min). We conclude that the mechanism of low calcium mediated loss of endothelial barrier function is mediated in part by a PKC dependent endocytosis of endothelial cadherins, which may involve interactions with the actin cytoskeleton. Physiological regulation of the in vivo endothelial barrier may also involve PKC dependent-actin mediated endocytosis of cadherin junctional elements.

Role of cadherin internalization in hydrogen peroxide-mediated endothelial permeability.

Exposure of endothelial monolayers to hydrogen peroxide results in increased solute permeability in a time- and dose-dependent fashion. This effect is prevented by either staurosporine, an inhibitor of PKC, or by Gö6976, an inhibitor of "classical" PKC isoforms. Immunohistochemistry of peroxide-treated monolayers illustrates a loss of cadherin staining at cell junctions and gap formation predominantly at tri-cellular junctions. Both staurosporine and Gö6976 prevented peroxide-induced gap formation. Peroxide also stimulated internalization of cadherins as measured by the trypsin protection assay, which was not blocked by staurosporine or Gö6976. These data suggest that peroxide causes: 1) a time- and dose-dependent increase in permeability and dose-dependent increase in gap formation, both of which are PKC dependent; and 2) promotes PKC-independent cadherin internalization. These data indicate that cadherin internalization may be part of the mechanism through which oxidants regulate solute permeability.

Vascular permeability factor/vascular endothelial cell growth factor-mediated permeability occurs through disorganization of endothelial junctional proteins.

Vascular permeability factor/vascular endothelial growth factor stimulates endothelial proliferation, angiogenesis, and increased vascular permeability in vivo. We investigated mechanisms of vascular permeability factor-mediated endothelial monolayer permeability changes in vitro. [14C]Albumin flux across endothelial monolayers was measured following a 90-min exposure to vascular permeability factor (660 pM). Vascular permeability factor increased albumin flux to 3.4 times that of control albumin flux. Endothelial monolayers were also incubated for 90 min with vascular permeability factor plus Gö6976 (10 nM), staurosporine (1 microM), wortmannin (10 nM), AG126 (1 and 2.67 microM), and PD98059 (20 microM). Vascular permeability factor-mediated permeability was not blocked by Gö6976, an antagonist of "classical" protein kinase C, staurosporine, a pan-protein kinase C antagonist, nor wortmannin, a PI3-kinase blocker, but was blocked by incubation with AG126 or PD98059, inhibitors of mitogen-activated protein kinase activation. Immunofluorescent staining of the junctional proteins VE-cadherin and occludin showed a loss of these proteins from the endothelial junction that was prevented by co-incubation with AG126 or PD98059. These data demonstrate that vascular permeability factor increases albumin permeability across endothelial monolayers in vitro and suggests that permeability increases through rearrangement of endothelial junctional proteins involving the mitogen-activated protein kinase signal transduction pathway.