| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Pharmacology, University of Illinois, College of Medicine, Chicago, Illinois
Submitted 17 May 2004 ; accepted in final form 24 March 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
morphometric analysis; basal lung vascular permeability; interendothelial junctions; nitric oxide synthase inhibition; endothelial nitric oxide synthase knockout mice; transendothelial transport
NO, a diatomic free radical with a half-life of a few seconds, is produced by eNOS in different cell types, most prominently ECs (24, 41, 42). The EC-generated NO controls the basal tone of arterioles and venules (67) and serves to prevent inappropriate platelet aggregation (19) and neutrophil adhesion to ECs (14). However, the role of constitutively generated NO in regulating vascular permeability is still a matter of debate. In some studies NO was shown to increase permeability in vascular beds (7, 8, 17, 27–30, 40), whereas it decreased vascular permeability in other studies (31, 35–37, 52, 70). Studies also showed that NOS inhibitors increased vascular permeability (1, 13, 32), raising the possibility that NO production may tonically regulate endothelial barrier function. In the present study, based on morphological observations of microvessels in multiple vascular beds, including the pulmonary circulation, we show that the absence of NO [as induced by eNOS inhibition with N-nitro-L-arginine methyl ester (L-NAME) or eNOS gene deletion in mice] causes the opening of IEJs in the capillary and venular segments of lung and other vascular beds. Structural alterations in the junctions were coupled to increased vascular permeability to tracer albumin. Thus our study demonstrates that eNOS-derived NO constitutively regulates basal vascular permeability.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Wild-type (WT) C57BL/6, eNOS–/– (strain B6.129P2-Nos3tm1Unc), and inducible nitric oxide synthase knockout (iNOS–/–, strain B6.129P2-Nos2tm1lan) mice were purchased from Jackson Labs (Bar Harbor, ME). Breeding colonies were maintained in the University Animal Facility. All experiments were made in accordance with policies of the institutional Animal Care Committee. All mice were fed a normal diet.
Reagents
We purchased from Sigma Chemical (St. Louis, MO) leupeptin, glycerol, benzamidine, phenylmethylsulfonyl fluoride (PMSF), 3-aminopropanolthriethoxysilane, 3% Monastral blue solution and Monastral blue, bovine serum albumin (BSA), lanthanum chloride (LaCl), L-NAME, and glycerol. We purchased from Amersham Pharmacia (Piscataway, NJ) Triton X-100, Nonidet P-40 (NP-40), sodium dodecyl sulfate (SDS), Tween 20, nitrocellulose (NC) membranes, and all chemicals for electrophoresis. We purchased from ElectronMicroscopy Science (Fort Washington, PA) paraformaldehyde (PFA), glutaraldehyde (GA), lanthanum nitrate, Surmount water base, Entellan, polyvinylpyrrolidone (PVP), and all electron microscopy (EM)-grade reagents. Protein A/G immunobeads were purchased from Calbiochem (San Diego, CA), and Vectashield mounting medium was from Vector Lab (Burlingame, CA). Bicinchoninic acid (BCA) kit and enhanced chemiluminescence (ECL) kit were from Pierce (Rockford, IL). Anti-eNOS and anti-iNOS antibodies (Ab) and horseradish peroxidase (HRP)-conjugated affinity-purified goat anti-rabbit and anti-mouse immunoglobulin G (IgG) were purchased from Transduction Laboratories (San Diego, CA). Rabbit anti-dinitrophenylated albumin (A-DNP) Ab and goat anti-rabbit HRP-coupled Ab were from Chemicon (Temecula, CA).
A-DNP, the albumin tracer used in these studies, was prepared as described in (44). Homogeneity, presence of BSA aggregates, and conformational changes induced by the derivatization procedure were checked for every plot. Only monomeric A-DNP, which is an adequate substitute for the native albumin, was used in these experiments.
Biochemical Procedures
Tissue lysates were prepared at 4°C using a solubilization buffer containing 0.5% SDS + 1% NP-40 in phosphate buffered-saline (PBS), plus protease inhibitors (1 mM leupeptin, 2 mM benzamidine, 1 mM PMSF, and 2 mM Na-pyrophosphate). Organs free of blood after perfusion with cold buffer (PBS or Hanks') containing 1% glucose and mixture of protease inhibitors were hand-minced, suspended in solubilization buffer at a ratio of 1:9 (wt/vol), and homogenized in a Warring blender set at maximum speed (4°C). After 1 h of being stirred at room temperature (RT), the resulting tissue lysates were clarified by centrifugation for 1 h at 40,000 rpm at 4°C in a Beckman Optima TLX ultracentrifuge with the TLA-50 rotor.
Protein concentrations were determined using the BCA method (61), with the Pierce kit and BSA as standard.
Immunoblotting. Total protein (200 µg) from tissue lysates was loaded on a preparative minigel (1.5 mm, 5–20% SDS-PAGE gradient) run at 150 V, and transferred to NC as in Ref. 63 using an Idea Scientific blotting apparatus at 4°C and 750 mA of constant current for 90 min. Strips of NC membranes containing the transferred proteins were blocked for 2 h at RT with BSA 3% + 0.05% Tween 20 + 1% goat serum in PBS and incubated for 1 h at RT with anti-eNOS or anti-iNOS Ab diluted 1:1,000 in the same blocking buffer. NC membranes were washed (3 x 10 min) with washing buffer (PBS + 1% goat serum + 0.2% Triton X-100) and incubated for 1 h at RT with the reporter Ab (HRP-coupled affinity purified goat anti-rabbit or anti-mouse IgG). After another wash, the reaction was visualized using an ECL kit and X-Omat Blue film.
ELISA. Enzyme-linked immunoassays were carried out as in Ref. 48 using 100 µg/ml (total protein concentration) as starting concentration and HRP-conjugated affinity-purified goat anti-rabbit IgG as the reporter Ab. The reaction was developed with 3,3',5,5'-tetramethylbenzidine and read at 450 nm in a Molecular Dynamics ELISA reader. In brief, to determine transendothelial albumin transport, the albumin tracer A-DNP was allowed to interact for 10 min with the vascular beds, after which they were flushed free of tracer, organs were removed, cleared of connective tissue, weighed, hand-minced, and homogenized with a Polytron at a 1:10 (wt/vol) ratio in PBS. The ensuing homogenates were centrifuged at 150,000 g in a Beckman TLX tabletop ultracentrifuge for 1 h using the TLA 55 rotor. The pellets were discarded and supernatants (representing the diluted interstitial fluid which included the tracer) were used to determine the total protein content and the concentration of A-DNP. The amount of transported tracer was quantified using A-DNP as standard. To calculate the concentrations of A-DNP in the final supernatants, a standard curve was generated using known concentrations of A-DNP tracer, and the amount of tracer transported was determined at a series of decreasing concentrations on the linear part of the standard curve. The amount of tracer transported (expressed as ng A-DNP transported per mg tissue wet weight over 5 min) was used to assess vascular permeability to albumin in different organs.
NO Measurement. Blood collected by cardiac puncture was centrifuged, and serum concentrations of the NO metabolites, nitrite and nitrate, were measured using an NO analyzer (ENO-20, Eicom) as in Ref. 23.
Tissue Experiments
Effects of NOS inhibition using L-NAME. Studies were initially carried out at the light microscopic level using the mouse cremaster preparation as described in Ref. 34 and a diaphragmatic muscle preparation by essentially the same technique. Subsequent studies employing detailed EM analysis addressed alterations in permeability of multiple vascular beds, including the lung vasculature. NOS activity was inhibited in two ways: 1) direct administration of a bolus dose of 30 mg/kg L-NAME iv with the mice studied in the first 4 h or 2) chronic administration of L-NAME (for 7–60 days) added to drinking water; intake of inhibitor was 30–40 mg/day as in Ref. 51. The control groups received 0.2 ml of sterile saline solution administered intravenously for acute treatment and regular water for the chronic inhibition group. Systolic blood pressure measured by the tail-cuff method (26), and body weights were recorded biweekly. In the chronic inhibition group, the mice were used for morphological and biochemical analyses at the eighth week of treatment.
Tracer experiments. Mouse perfusions were done under anesthesia induced with ketamine-xylazine (10 mg/ml ketamine + 0.5 mg/ml xylazine) injected intraperitoneally as in Ref. 46. For perfusion of all nonvascular beds, the abdominal aorta was exposed; an inlet catheter was inserted above the iliac artery bifurcation and secured in place while another catheter was inserted in the vena cava caudalis and used as outlet. For lung perfusion, the inlet catheter was inserted into pulmonary artery with the outflow catheter in the left atrium. The vasculature, via either route, was flushed blood-free for 3 min with oxygenated Hanks' solution warmed up to 37°C, supplemented with 4 g/l glucose, and perfused with a peristaltic pump (Pharmacia P1) at a flow rate of 3 ml/min. The A-DNP was added to the perfusate at a concentration of 5 mg/ml and presented to the vascular beds studied for 5 or 10 min (10, 11). The unbound tracer was flushed from the vasculature by 3 min perfusion with Hanks' solution. The organs were then fixed by in situ perfusion of a fixative mixture (4% PFA + 2% GA + 1 mM CaCl2 in cacodylate buffer, pH 7.4) for 10 min. The lungs, heart, diaphragm, and skeletal muscle were removed, cut into 3 x 3 x 3-mm sections, and processed for morphological surveys. For assessment of the permeability of different vascular beds, the organs were flushed blood free, weighed, and then homogenized as described above. The final supernatants of these homogenates were used in ELISA to quantify vascular permeability as described (44, 48).
Microscopy and Immunocytochemistry
Light microscopic studies. eNOS–/– and L-NAME-treated mice were injected intravenously with a 0.2-ml solution of the Monastral blue dye. At different time points, cremaster and diaphragm muscles were fixed in situ by injecting 5 ml of freshly prepared 8% PFA in 0.1 M Na-cacodylate buffer (pH 7.2) intraperitoneally for cremaster and 3 ml of the same fixative into the thoracic and abdominal cavities for the diaphragm. After 15 min at RT, the muscles were removed, dissected free of connective tissue, and postfixed in 2% PFA + 0.5% GA + 2.5% PVP in 0.1 M Na-cacodylate for 24 h at RT. Selected specimens (chosen under a stereomicroscope) were cleared for 96 h at RT in glycerol and mounted with Entellan. Micrographs were obtained with a Zeiss Axiophot 2 microscope.
EM studies.
The tissues were fixed in situ by perfusing a fixative mixture (3% PFA + 2.5% GA + 1% LaCl in 0.1 PIPES, pH 7.2) for 15 min, and selected specimens were further fixed with the same mixture (without LaCl) for 1 h at RT. All specimens were postfixed in 2% OsO4 in acetate veronal buffer, pH 6.8, for 1 h on ice, stained in the dark for 1 h with 7.5% uranyl-magnesium acetate (UA), dehydrated through increasing concentrations of ethanol (50, 70, 90, 100%), then in propylene oxide, and embedded in Epon 812. Specimens embedded in Epon were cured for 72 h at 90° C, and sections 
60 nm thick obtained with a Leica microtome were examined and photographed in a Philips TM-10 electron microscope. For routine examinations, sections from WT and eNOS–/– mice were stained with 7.5% UA for 5 min and saturated lead citrate for 3 min and then examined. When LaCl was added to the fixative as a tracer, the tissue samples were not stained with UA and lead citrate.
Immunocytochemistry at the EM level.
The specimens were collected from tissues perfused with A-DNP as described above. Collected specimens were further fixed by immersion in a triple fixative (58) for 1 h on ice, stained in block for 30 min with 7.5% UA, dehydrated slowly in graded ethanol solutions, and embedded in Epon. A-DNP was detected by a postembedding immunostaining procedure (10, 11). In brief, sections of 50–60 nm obtained from blocks of embedded tissues were quenched for 45 min at RT in PBS + 1% BSA (A-PBS), incubated for 2 h at RT with anti-DNP IgG diluted 1:2,000 in A-PBS, washed 3 x 15 min with A-PBS, incubated with gold-tagged (6 nm) anti-rabbit IgG diluted 1:5,000 in A-PBS, fixed shortly (5 min) with 2.5% GA, examined, and micrographed in a Jeol 1220 electron microscope operated at 80 kV.
Morphometric Analysis
Light microscopy. Sections (on average 30 slides for every organ and each condition; i.e., WT, L-NAME-treated, and eNOS–/– mice) from lung, heart, skeletal muscle, diaphragm, and gut were analyzed using the ImagePro Morphometric program, which allowed us to perform automated analyses of the number of vessels in reference to the total area of section, mean vascular diameter, number of histochemical-positive ("tattooed") vessels. The images were acquired using an Axioplan 2 Carl Zeiss microscope equipped with a color digital camera, and the mentioned morphometric program was run on an Apple G4 computer.
EM. We carried out extensive morphometric analysis of lungs from WT mice (C57BL) and eNOS–/– mice. For lungs, four to six Epon blocks were used for thin sectioning, and six grids per block (every grid with 15–25 sections) were examined. When lung tissues were used for morphometric analysis, vertical uniform random and isotropic uniform random sections were prepared from each left and right lung as in Ref. 16. In brief, the vertical axis of each lung was identified, and 4-mm-thick slices were cut perpendicularly to the axis starting from a randomly chosen point. The slices were cut into bars at 3-mm intervals, and every fifth tissue block was selected from a randomly chosen starting point. Sections (60–70 nm) were obtained at random, and only vessels with a full circumference on the grid mesh were photographed. The images were acquired with a Gatan charge-coupled device camera coupled to an Apple G4 computer. A total of 255 images for each part of a vascular bed (arterial, capillary, and venular) and each organ studied at the final magnification of x28,000 were stacked as a queue, and grid no. 3 from the Stereology Toolbox program (Morphometrix) was used to quantify the main endothelial features.
Statistical Analysis
Data are expressed as means ± SD. Statistical analysis was performed using one-way analysis of variance and Student's t-test with a post hoc Bonferroni correction for multiple comparisons performed to identify group differences accounted for the significant overall ANOVA. Significance was set at P < 0.001.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We compared eNOS–/– and iNOS–/– mice for the relative expressions of eNOS and iNOS. Immunoblotting using anti-eNOS Ab failed to detect the enzyme in any organ of eNOS–/– mice, whereas there was iNOS immunoreactivity in these mice; the opposite was the case for eNOS–/– mice (Fig. 1). We also determined mean arterial pressure with the tail-cuff method in L-NAME-treated and eNOS–/– mice. Arterial pressure averaged 137 ± 22 mmHg (n = 47) in L-NAME-treated mice and 135 ± 24 mmHg (n = 172) in eNOS–/– mice, compared with the value of 102 ± 29 mmHg in WT mice (n = 66) (P < 0.001). Blood NO level (expressed as NOx), which under the basal condition reflects the eNOS-derived production, was threefold less in L-NAME-treated and eNOS–/– mice than WT mice (Table 1). Based on these results, we investigated the vascular permeability characteristics of both L-NAME-treated and eNOS–/– mice.
|
|
In the initial studies carried out at the light microscopic level, we observed extensive tattooing1 (Monastral blue deposits) of postcapillary venules (10–40 µm diameter), muscular venules (40–60 µm diameter), and venules (>70 µm diameter) in cremaster and diaphragm muscles of L-NAME-treated and eNOS–/– mice as illustrated in Fig. 2, A–D. This finding was indicative of extensive dye leakage into the extravascular space. In WT mice, we observed no deposits of the Monastral blue dye. We also noted differences in density and frequency of tattooing in different vascular segments of L-NAME-treated and eNOS–/– mice. Measurement of tattooed areas in 24 preparations from cremaster and 42 from diaphragm showed that the surface area of affected fields varied from <0.5 µm2 in regions where capillaries reside to 1.5 µm2 in postcapillary venules and to 2–3 µm2 in collecting venules. There was no significant difference between the effects of acute vs. chronic NOS inhibition with L-NAME (42% of total number of vessels were affected after acute L-NAME administration vs. 38% after chronic L-NAME administration). In the acute NOS inhibition protocol, open IEJs with extravascular deposits of the tracer were seen in the first minute after application of L-NAME and reached a plateau at 30 min after L-NAME administration.
|
Morphological Analysis of IEJs at the EM Level
We observed upon extensive EM analysis of vascular beds from different organs open IEJs in vessels of L-NAME-treated mice as illustrated in Figs. 3 and 4. Figure 3 shows open IEJs from the murine lung vascular bed in a postcapillary venule (Fig. 3A), large muscular vein (Fig. 3B), and capillary (Fig. 3C). Insets in Fig. 3, B and C, show normal IEJ spaces (sealed by tight junction) from WT mice. The same findings obtained in other vascular beds are illustrated in Fig. 4. IEJs were open in heart capillaries (Fig. 4A), diaphragm capillaries (Fig. 4B), and mesenteric venules (Fig. 4C). Note a platelet migrating outward through the open IEJ of this mesenteric vessel (Fig. 4C). Insets in Fig. 4, B and C, illustrate normal IEJs from these vascular segments in control mice (with the interendothelial spaces sealed by tight junctions that are the most apical IEJ structure).
|
|
|
|
|
|
|
|
Based on these morphological findings, we conclude that: 1) tattooing of vessel walls induced by L-NAME observed by light microscopy (Fig. 2) is a reflection of open IEJs and of accumulation of tracer particles in the perivascular space, 2) capillaries, postcapillary venules, and muscular venules are the main segments affected in all vascular beds by the removal of NO, and 3) open IEJs are the essential hallmark of removing NO by pharmacological inhibition of NOS activity or genetic deletion of eNOS, in all vascular beds examined.
Transendothelial Albumin Permeability
We next used the A-DNP as a tracer to quantify changes in albumin transport due to the observed alteration of IEJs. The pathways followed by A-DNP from vascular lumen into the pericapillary space were identified by immunocytochemistry as in Refs. 11 and 44, and the permeability for A-DNP was quantified as described in METHODS. Figure 11 demonstrates the interactions of A-DNP (5 mg/ml) with the endothelial barrier when presented by perfusion. Figure 11, A and B, shows the A-DNP labeling pattern in ECs from the venular end of a lung microvessel in WT mice. Figure 11C illustrates the interactions of A-DNP with the ECs in venular microvessels of lungs from eNOS–/– mice. The same findings were true for the labeling of lung capillaries as well as of microvessels from all organs studied. The IEJs are open and penetrated by the tracer particles when NO is removed from a vascular bed. In WT mice, under basal conditions, the tracer 1) binds to luminal surface, 2) labels all caveolae open at the luminal side, most of the caveolae free inside the cell, and caveolae discharging their content on the abluminal side, 3) is detected at sites of discharging caveolae, and 4) fails to penetrate the IEJ space. The major findings in the lung vascular bed of eNOS–/– mice are open IEJs penetrated by tracer particles (Fig. 11C). The same findings were evident in L-NAME-treated mice injected intravenously with A-DNP (data not shown) and in all other vascular beds studied.
|
|
The morphometric analysis of murine lung vascular bed is summarized in Table 3. The results indicate an opening of 34% of IEJs in capillaries and 48% of IEJs in venules of mice lacking NO, whereas in WT mice no open IEJs were found in capillaries and only 12% of venular IEJs were open. We found extremely rare (<0.5%) IEJs open on the arterial end of eNOS–/– mice. The main structural characteristics of caveolae (average diameter of 70 nm, their surface and volume densities) open to the luminal or abluminal side or apparently free in the cytosol in both L-NAME-treated and eNOS–/– mice were within the normal values published (56, 62). Most of the parameters shown in Table 3 were also in the normal range (56, 68).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In both eNOS–/– and L-NAME-treated mice, the basal plasma levels of NO were low, consistent with the fact that NO is released constitutively by eNOS. In the present study, we reassessed the role of NO constitutively released by eNOS in tonically regulating basal endothelial permeability. We evaluated the morphological alterations of continuous type of endothelia from multiple vascular beds including the lung, and in addition we also measured endothelial permeability using A-DNP as a monomeric albumin tracer (11, 44) in both eNOS–/– and L-NAME-treated mice. We were also able to map out the routes of albumin transport in microvessels from the same vascular beds, in the same experimental conditions by immunocytochemistry at the EM level.
We detected, in the initial series of studies at the light microscopic level, the leakage of Monastral blue dye from cremaster and diaphragmatic vascular bed of both eNOS–/– and L-NAME-treated mice. This was not seen in microvessels from control mice. This dye, with molecular dimensions ranging from 0.02–1 µm, does not normally permeate the endothelial barrier (20). However, upon removal of NO (by L-NAME treatment or deletion of eNOS in eNOS–/– mice), the leakage of Monastral blue was similar to the effects seen with permeability-increasing mediators such as platelet-activating factor (45) and histamine (34, 69). However, unlike these mediators, which induce junctional opening within minutes, the increase in permeability produced by L-NAME was delayed, reaching a plateau at 30 min after treatment. This finding suggests that different molecular mechanisms may account for the effect obtained after blocking NO production. Although the basis for the leakage of Monastral blue from venular segments was not evident from the light microscopy observations, one possibility is the formation of minute gaps between ECs at the level of IEJs (38).
To address cellular basis for the effect of removal of NO in increasing the vascular leakage of Monastral blue dye, we carried out a comprehensive EM analysis of the endothelial barrier under basal conditions in WT mice and of alterations induced by either L-NAME-treatment or eNOS deficiency. When the general morphological characteristics of mouse lung vasculature were analyzed (WT vs. L-NAME-treated and eNOS–/– mice), the only statistically significant difference was the invariable presence of open IEJs in the mice lacking NO. This effect was seen in IEJs of capillaries and venules of the multiple vascular beds examined including the pulmonary circulation, but it was rarely seen in the arterial segments. This finding points to the key role of eNOS-derived NO as a determinant factor of basal endothelial permeability via its action in maintaining the integrity of IEJs.
We also used the electron-opaque tracer LaCl, which was presented to different vascular beds at the same time as the fixative to visualize the pathways and structures involved in the transport of the tracer from the vascular lumina and to resolve the dimensions of IEJ openings. Our results showed that in the absence of NO the tracer was cleared from the vascular lumina mainly via open IEJs. We also showed with another tracer, the monomeric A-DNP, that in the absence of NO this tracer leaked across the endothelial barrier via open IEJs while it is still transported by caveolae. In contrast, in control vessels, albumin did not penetrate the IEJs, and its transport occurred solely via caveolae as demonstrated in previous studies (464748, 54–59). Our serial sectioning data showing that IEJs of continuous endothelia are normally impermeable to LaCl are in agreement with Wagner and Chen (65, 66), who, using the same technology applied on terbium-labeled microvessels, demonstrated that, under basal conditions, endothelial caveolae were the only subcellular structures participating in transendothelial transport. In control vessels, the LaCl or A-DNP tracer failed to penetrate IEJs beyond the tight junctions of lung vascular endothelia and the other vascular beds studied.
Using the A-DNP as tracer we determined that lung permeability for albumin increased by 35–40% in mice lacking NO (eNOS–/– and L-NAME treated) compared with WT mice. This augmented permeability can be explained by the increase in the number of leaky venular IEJs (from 12% under basal condition in WT mice to 45% in the absence of NO) and by an increased number of open junctions in capillaries (from 0 under basal condition to 34% in the absence of NO). The presence of open IEJs in mouse lung venules under basal condition is essentially similar to the observation in rat diaphragm and heart postcapillary venules made by Simionescu et al. (59). Thus our observations provide a strong argument for the crucial role of NO as a determinant factor of basal vascular permeability by regulating capillary and venular endothelial junctional integrity and by controlling the dimensions of IEJs. NO acts primarily at the level of the capillary and venular IEJs, which are tight enough to exclude molecules with molecular diameters 
2–3 nm (48); therefore, albumin, because of its molecular dimensions (4 x 4 x 15 nm), never crosses the endothelial barrier via IEJs under basal conditions. However, in the absence of NO (as in eNOS–/– and L-NAME-treated mice), the IEJs are opened (>12 nm) and the transport of albumin mainly takes place via this modified paracellular pathway.
The present results should be contrasted with data showing microvascular hyperpermeability when excessive amounts of NO are present in the vasculature (47) or when caveolin-1 is deleted in mice by homologous recombination (53) and large amounts of eNOS-derived NO are produced by ECs. Previous studies have shown that increased levels of NO such as induced by VEGF, which activates eNOS by Ca2+/calmodulin and Akt-mediated phosphorylation of eNOS (3, 4, 9, 33), promote the leakiness of the endothelial barrier (6, 22, 60), at least partially by opening IEJs. As caveolin-1 functions as an inhibitor of eNOS activity (5, 12), the increase in vascular permeability in caveolin-1–/– mice was ascribed to the increased production of NO (53) and its prolonged half-life in the circulation (47). Thus open IEJs are also found when excessive production of NO as in inflammatory states (18) or when a more stable form of NO [NO attached to albumin by nitration or nitrosilation (47)] is present in circulation, and as a consequence its excess disturbs the paracellular junctional pathway (47, 64). This alteration of the endothelial paracellular pathway generates the recorded increase in vascular permeability. The increase in endothelial permeability in the above instances was prevented, to different degrees, by the NOS inhibitor L-NAME, indicating that the high levels of NO were responsible for the increase. However, it is likely that the molecular mechanisms responsible for the opening of IEJs when ECs are exposed to large amounts of NO are different from the increased endothelial permeability induced by the absence of NO as demonstrated by the present study. The molecular basis for the IEJ effect induced by the absence of NO remains unclear.
High levels of NO may modify actin polymerization and promote endothelial shape change, resulting in increased junctional permeability (2), whereas it is possible that basal production of eNOS-derived NO serves to stabilize IEJs. Studies have shown that NO induces the formation of endothelial gap junctions by promoting the incorporation of connexin-40 into the plasma membrane secondary to activation of protein kinase A (15); thus NO-mediated gap junction formation may favor cell-cell contact at the level of IEJs. Another possibility is that the constitutive production of NO may regulate the distribution of cortical actin and its interaction with endothelial junctions and thus contributes to maintenance of IEJ integrity as in the case of sphingosine-1-phosphate (39).
The question why eNOS–/– mice are viable and their tissues are not grossly edematous, despite the leaky IEJs, is unresolved. It is possible compensatory mechanisms preventing edema formation are activated in these mice. These mechanisms may include an increase in the lung lymphatic drainage system (43) and an increase in interstitial hydrostatic pressure. Both of these would serve to limit or even resolve tissue edema formation in the unstressed mouse in the face of increased endothelial permeability induced by absence of NO.
In summary, the findings that open IEJs are the hallmark of NO deficiency as well as of NO excess give us the confidence to conclude that a restrictive endothelial barrier, in a given vascular bed, requires a physiological concentration of NO.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 The term "tattooing" is used to describe tracer deposition (Monastral blue) in the microvessel wall as a consequence of open IEJs and trapping of tracer particles in the perivascular space; see Majno et al., J Biophys Biochem Cytol 11: 607–611, 1961, and Predescu et al., Eur J Cell Biol 69: 86–98, 1996. 
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Nakagawa, T. Kosugi, M. Haneda, C. J. Rivard, and D. A. Long Abnormal Angiogenesis in Diabetic Nephropathy Diabetes, July 1, 2009; 58(7): 1471 - 1478. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Wojciak-Stothard, B. Torondel, L. Zhao, T. Renne, and J. M. Leiper Modulation of Rac1 Activity by ADMA/DDAH Regulates Pulmonary Endothelial Barrier Function Mol. Biol. Cell, January 1, 2009; 20(1): 33 - 42. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Dossumbekova, E. V. Berdyshev, I. Gorshkova, Z. Shao, C. Li, P. Long, A. Joshi, V. Natarajan, and T. L. Vanden Hoek Akt activates NOS3 and separately restores barrier integrity in H2O2-stressed human cardiac microvascular endothelium Am J Physiol Heart Circ Physiol, December 1, 2008; 295(6): H2417 - H2426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Surapisitchat, K.-I. Jeon, C. Yan, and J. A. Beavo Differential Regulation of Endothelial Cell Permeability by cGMP via Phosphodiesterases 2 and 3 Circ. Res., October 12, 2007; 101(8): 811 - 818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Predescu, D. N. Predescu, and A. B. Malik Molecular determinants of endothelial transcytosis and their role in endothelial permeability Am J Physiol Lung Cell Mol Physiol, October 1, 2007; 293(4): L823 - L842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Van Driessche, J. L. Kreindler, A. B. Malik, S. Margulies, S. A. Lewis, and K.-J. Kim Interrelations/cross talk between transcellular transport function and paracellular tight junctional properties in lung epithelial and endothelial barriers Am J Physiol Lung Cell Mol Physiol, September 1, 2007; 293(3): L520 - L524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-L. Yeh, D.-Y. Lu, C.-J. Lin, H.-C. Liou, and W.-M. Fu Inhibition of Hypoxia-Induced Increase of Blood-Brain Barrier Permeability by YC-1 through the Antagonism of HIF-1{alpha} Accumulation and VEGF Expression Mol. Pharmacol., August 1, 2007; 72(2): 440 - 449. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Scalia, Y. Gong, B. Berzins, L. J. Zhao, and K. Sharma Hyperglycemia Is a Major Determinant of Albumin Permeability in Diabetic Microcirculation: The Role of {micro}-Calpain Diabetes, July 1, 2007; 56(7): 1842 - 1849. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Song, S. Ge, and J. S. Pachter Caveolin-1 regulates expression of junction-associated proteins in brain microvascular endothelial cells Blood, February 15, 2007; 109(4): 1515 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Zhao, S. Wang, H. Cheng, M.-z. Zhang, T. Takahashi, A. B. Fogo, M. D. Breyer, and R. C. Harris Endothelial Nitric Oxide Synthase Deficiency Produces Accelerated Nephropathy in Diabetic Mice J. Am. Soc. Nephrol., October 1, 2006; 17(10): 2664 - 2669. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Y. Yuan New insights into eNOS signaling in microvascular permeability Am J Physiol Heart Circ Physiol, September 1, 2006; 291(3): H1029 - H1031. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hatakeyama, P. J. Pappas, R. W. Hobson II, M. P. Boric, W. C. Sessa, and W. N. Duran Endothelial nitric oxide synthase regulates microvascular hyperpermeability in vivo J. Physiol., July 1, 2006; 574(1): 275 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Wagner, G. Karagulova, J. Jenkins, J. Bishai, and J. McClintock Changes in lung permeability after chronic pulmonary artery obstruction J Appl Physiol, April 1, 2006; 100(4): 1224 - 1229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mukherjee, M. Tessema, and A. Wandinger-Ness Vesicular Trafficking of Tyrosine Kinase Receptors and Associated Proteins in the Regulation of Signaling and Vascular Function Circ. Res., March 31, 2006; 98(6): 743 - 756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Miyawaki-Shimizu, D. Predescu, J. Shimizu, M. Broman, S. Predescu, and A. B. Malik siRNA-induced caveolin-1 knockdown in mice increases lung vascular permeability via the junctional pathway Am J Physiol Lung Cell Mol Physiol, February 1, 2006; 290(2): L405 - L413. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
