Published online before print August 17, 2007, doi: 10.1161/CIRCRESAHA.107.154229
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2007 American Heart Association, Inc.
Cellular Biology |
Differential Regulation of Endothelial Cell Permeability by cGMP via Phosphodiesterases 2 and 3
From the Department of Pharmacology (J.S., J.A.B.), University of Washington School of Medicine, Seattle; and Cardiovascular Research Institute (K.-I.J., C.Y.), University of Rochester, NY.
Correspondence to Joseph A. Beavo, PhD, University of Washington, Department of Pharmacology, 1959 NE Pacific St, Box 357280, Seattle, WA 98195-7280. E-mail beavo{at}u.washington.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Endothelial barrier dysfunction leading to increased permeability and vascular leakage is an underlying cause of several pathological conditions, including edema and sepsis. Whereas cAMP has been shown to decrease endothelial permeability, the role of cGMP is controversial. Endothelial cells express cGMP-inhibited phosphodiesterase (PDE)3A and cGMP-stimulated PDE2A. Thus we hypothesized that the effect of cGMP on endothelial permeability is dependent on the concentration of cGMP present and on the relative expression levels of PDE2A and PDE3A. When cAMP synthesis was slightly elevated with a submaximal concentration of 7-deacetyl-7-(O-[N-methylpiperazino]-
-butyryl)-dihydrochloride–forskolin (MPB–forskolin), we found that low doses of either atrial natriuretic peptide (ANP) or NO donors potentiated the inhibitory effects of MPB–forskolin on thrombin-induced permeability. However, this inhibitory effect of forskolin was reversed at higher doses of ANP or NO. These data suggest that cGMP at lower concentrations inhibits PDE3A and thereby increases a local pool of cAMP, whereas higher concentrations cGMP activates PDE2A, reversing the effect. Inhibitors of PDE3A mimicked the effect of low-dose ANP on thrombin-induced permeability, and inhibition of PDE2A reversed the stimulation of permeability seen with higher doses of ANP. Finally, increasing PDE2A expression with tumor necrosis factor-
reversed the inhibition of permeability caused by low doses of ANP. As predicted, the effect of tumor necrosis factor- on permeability was reversed by a PDE2A inhibitor. These findings suggest that the effect of increasing concentrations of cGMP on endothelial permeability is biphasic, which, in large part, is attributable to the relative amounts of PDE2A and PDE3A in endothelial cells.
Key Words: phosphodiesterase • cGMP • endothelial permeability • TNF-
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The endothelium plays an important role in maintaining normal vascular function. In addition to maintaining a nonthrombogenic surface, secreting anticoagulation factors, responding to and participating in inflammatory signaling and regulating vascular tone, the endothelium also acts as an active barrier between the blood and the underlying tissue. Endothelial dysfunction caused by injury or inflammatory signals has been shown to lead to numerous pathological conditions. For example, inflammatory cytokines can induce adhesion molecule expression in endothelial cells, thereby promoting the adherence and migration of immune cells into the vessel wall, leading to atherosclerosis.1 Dysfunctional endothelium also leads to decreased barrier function or increased permeability. Increased endothelial permeability is characteristic of many diseases and pathological conditions, including atherosclerosis, asthma, tumor growth, edema, and sepsis.2,3
Several inflammatory mediators, including vascular endothelial growth factor, tumor necrosis factor (TNF)-, histamine, and thrombin, are known to increase endothelial permeability.2–4 TNF- increases permeability, at least in part, via activation of protein kinases such as Fyn and p38, whereas vascular endothelial growth factor, histamine, and thrombin all signal via increasing intracellular Ca2+. An increase in Ca2+ in endothelial cells leads to activation of myosin light chain kinase and endothelial cell contraction, resulting in decreased barrier function. Whereas Ca2+ increases endothelial cell permeability, the second messenger, cAMP, has, in general, been shown to decrease permeability.2 ß-Adrenergic agonists, adenylate cyclase activators such as forskolin, and cAMP-phosphodiesterase inhibitors all have been shown to improve endothelial barrier function basally and under stimulated conditions. Although it is has been shown that cAMP can decrease endothelial cell permeability, the effects of cGMP are controversial.2 Both barrier-enhancing and -impairing effects of cGMP have been reported.2,5,6 These differences most commonly have been attributed to the differences in the types of endothelial cells and vascular beds studied and/or to whether or not cGMP-dependent protein kinase (PKG) was expressed. In this study, we test another possible mechanism, namely that the effect of cGMP signaling on permeability is dependent on the relative expression of cGMP-regulated cAMP-hydrolyzing phosphodiesterases in the endothelial cells that are differentially sensitive to the concentration of cGMP that is generated.
Cyclic nucleotide phosphodiesterases (PDEs) play a critical role in regulating a variety of cellular functions. By catalyzing the breakdown of cyclic nucleotides, PDEs control the amplitude, duration, and compartmentalization of cyclic nucleotide signaling in the cell. The PDE superfamily consists of 11 different family members distinguished by their unique regulation, enzymatic characteristics, structure, and pharmacological inhibitory profiles, as well as by their tissue, cellular, and subcellular expression.7 Endothelial cells have been shown to express cGMP-stimulated PDE2, cGMP-inhibited PDE3, cAMP-specific PDE4, and cGMP-specific PDE5.8 Although the presence of these PDEs in most endothelial cells has been shown, their relative expression is variable depending on the endothelial cells being studied.8 Because endothelial cells express both cGMP-stimulated PDE2 and cGMP-inhibited PDE3, endothelial responses to increases in cGMP might be expected to vary because of the differential effects cGMP has on these PDEs. cGMP can activate PDE2 by binding to the regulatory domain of PDE2, thereby increasing its rate of hydrolysis of both cGMP and cAMP.7 On the other hand, cGMP also can act as a high-affinity competitive inhibitor of PDE3 because the Vmax for cGMP is much less than the Vmax for cAMP.7 Thus the outcome of cGMP signaling on endothelial functions may vary depending on the degree to which PDE2 is activated and/or PDE3 is inhibited.
In the present study, we tested the hypothesis that the effect of cGMP on thrombin-induced permeability may be attributable to its differential regulation of PDE2 and PDE3 on cAMP hydrolysis. Specifically, we examined whether cGMP can potentiate the inhibitory effect of forskolin on thrombin-induced permeability by inhibiting cAMP hydrolysis by PDE3 and/or reverse the effect of forskolin by activating PDE2 cAMP hydrolysis. We found that low doses of guanylyl cyclase agonists, such as atrial natriuretic peptide (ANP) or NO donors, potentiated the effects of forskolin via cGMP inhibition of PDE3, whereas higher doses reversed this potentiation by cGMP activation of PDE2. Furthermore, as TNF- induces PDE2 expression in endothelial cells,9 we show that TNF- also decreases PDE3A expression in endothelial cells. Increasing PDE2 and decreasing PDE3 expression reverses the response of endothelial cells to low concentrations of cGMP. These findings suggest that the result of cGMP signaling on endothelial cell permeability is highly dependent on the concentration of intracellular cGMP, at least in part, because of the relative expression and activity of PDE2 and PDE3 in a particular endothelial bed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Permeability Assays
Permeability assays were performed as previously described.10 Briefly,
1x105 human umbilical vein endothelial cells (HUVECs) were seeded on gelatin-coated transwell units (6.5-mm diameter, 3.0-µm pore size; Corning). After 48 hours, media±10 ng/mL TNF- was changed, and measurements were performed 24 hours later. Before agonist stimulation, cells were rinsed once with medium 199 containing 1% BSA, followed by incubation in medium 199+1% BSA for 45 minutes. PDE inhibitors or vehicle were then added for an additional 15 minutes before stimulation. Permeability was measured by adding 1 mg of fluorescein isothiocyanate–dextran (Mr 42 000) per milliliter, together with agonists to the upper chamber. After incubation for 30 minutes, 50 µL of sample from the lower compartment was diluted with 150 µL of PBS and measured for fluorescence at 520 nm when excited at 492 nm with a Beckman DTX880 Multimode Detector. Results are presented as either fold increase in permeability over nonstimulated control or as percentage of thrombin maximum. Paired t test was performed using Prism v4.00. An expanded Materials and Methods section is available in the online data supplement at http://circres.ahajournals.org.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cGMP Has a Biphasic Effect on cAMP Inhibition of Thrombin-Induced Permeability
The consequence of cGMP signaling on endothelial barrier function has been controversial, as both increases and decreases in endothelial cell permeability in response to cGMP have been reported.2 We first tested whether the guanylyl cyclase agonists, ANP (a particulate guanylyl cyclase agonist) and S-nitroso-N-acetylpenicillamine (SNAP) (a NO donor and soluble guanylyl cyclase agonist) would affect permeability. ANP alone at 0.3 and 30 nmol/L and SNAP at 0.001 and 10 µmol/L had no significant effect on basal permeability (Figure IA and IB in the online data supplement). We then investigated whether these agonists could have an effect on thrombin-induced permeability. A low dose of ANP (0.3 nmol/L) significantly inhibited thrombin-induced permeability (supplemental Figure IA). However, this effect was completely reversed at a higher dose of ANP. Similarly SNAP at 0.001 µmol/L had no effect on thrombin-induced permeability, whereas 10 µmol/L potentiated the effect of thrombin (supplemental Figure IB). Because the guanylyl cyclase activators had differential effects on thrombin-induced permeability and we have confirmed that cAMP inhibits thrombin-induced permeability (supplemental Figure IIIA and IIIB), we hypothesized that cGMP may be differentially regulating cAMP PDEs. To try to mimic the in vivo condition of a modest adenylyl cyclase tone and test this hypothesis, we first increased endothelial permeability with thrombin (1 U/mL) and a submaximal dose of the water-soluble adenylate cyclase activator 7-deacetyl-7-(O-[N-methylpiperazino]-
-butyryl)-dihydrochloride–forskolin (MPB–forskolin) (0.1 µmol/L) was then used to slightly increase intracellular cAMP and inhibit thrombin-induced permeability by 10% to 15% of the thrombin maximum (100%). In vivo, one would expect that the effects of PDE regulation would be more pronounced when a cAMP tone is present, for example, from PGE2 release. We then examined the ability of cGMP generated in response to ANP (Figure 1A) and SNAP (Figure 1B) to either potentiate or reverse this inhibitory effect. As shown in Figure 1A, low doses of ANP (0.003 to 0.3 nmol/L) potentiated the inhibitory effect of cAMP on thrombin-induced permeability. The potentiating effect of low doses of ANP was reversed at higher doses (3 to 300 nmol/L). These concentrations of ANP dose-dependently increased intracellular cGMP under similar conditions (supplemental Figure IV). Similar results were seen with SNAP, which increases cGMP via activation of soluble guanylyl cyclase (Figure 1B). Furthermore, the effect of SNAP was inhibited by oxadiazolo quinoxalin-1-one (1H-[1,2,4]Oxadiazole[4,3-a]quinoxalin-1-one), a specific soluble guanylyl cyclase inhibitor (Figure 1C). These results suggest that cGMP has a concentration-dependent biphasic effect on cAMP inhibition of thrombin-induced permeability.
|
0.0302 vs thrombin+MPB–forskolin alone. #P
PKG Does Not Mediate the Effect of cGMP on cAMP Inhibition of Thrombin-Induced Endothelial Cell Permeability
There are several known effector molecules for cGMP, including PKG, cyclic nucleotide–gated (CNG) ion channels, and PDEs. To determine whether or not PKG or CNG channels mediated the effects of cGMP on endothelial permeability, we used 2 cGMP analogs, 8-Br-cGMP and 8-pCPT-cGMP, known to activate PKG and CNG channels but not regulate either PDE2 or PDE3.11 Neither 8-Br-cGMP nor 8-pCPT-cGMP had any significant effect on cAMP inhibition of thrombin-induced permeability (Figure 2). Therefore, PKG and CNG channels are not likely to mediate the effects of cGMP on endothelial permeability. Furthermore, we found that 8-Br-cAMP, a cAMP analog shown to activate PKA, Epac (exchange proteins activated directly by cAMP), and CNG channels, potentiated the inhibitory actions of MPB–forskolin on thrombin-induced permeability. This indicates that cyclic nucleotide analogs can permeate HUVECs and further demonstrates the barrier-enhancing effects of cAMP. In contrast to others who have reported a lack of PKG expression in HUVECs,12 we found PKG expression, which can be activated by cGMP analogs, in early passage HUVECs, as isolated in our laboratory (supplemental Figure V). Together, these results suggest that the effect of cGMP on endothelial cell permeability is not likely to be via PKG.
|
Roles of PDE2 and PDE3 in Mediating the Effects of cGMP on cAMP Inhibition of Thrombin-Induced Permeability
Because cGMP has a biphasic effect on cAMP-inhibited thrombin-induced permeability and this effect is not likely to be mediated by PKG or CNG, we hypothesized that the biphasic effect of cGMP is mediated via the inhibition of PDE3 at low doses of cGMP, followed by activation of PDE2 at higher doses of cGMP, thus altering the levels of cAMP in the cell in opposite directions. To test this hypothesis, we used the PDE2- and PDE3-specific inhibitors Bay 60-7550 (PDE2) and trequinsin (PDE3). Trequinsin (30 nmol/L) alone mimicked the actions of 0.03 nmol/L ANP on cAMP inhibition of thrombin-induced permeability (Figure 3A). This demonstrates that inhibition of PDE3 can potentiate the effect of MPB–forskolin on thrombin-induced permeability and suggests that the mechanism of low concentrations of cGMP signaling may be via PDE3 inhibition. If the effect of low concentrations of cGMP were attributable to a mechanism other than PDE3 inhibition, we might have expected an additive or synergistic effect with simultaneous addition of low concentrations of ANP with PDE3 inhibitors. When HUVECs were pretreated with trequinsin before addition of ANP, there was no additional potentiation of ANP on cAMP inhibition of thrombin-induced permeability (Figure 3A). Similar results were seen with cilostamide, a structurally different but selective PDE3 inhibitor (supplemental Figure VI). This suggests that the mechanism by which low doses of ANP improves barrier function is via cGMP inhibition of PDE3.
|
In other experiments, the PDE2 inhibitor Bay 60-7550 (100 nmol/L) blocked the actions of higher (30 nmol/L) concentrations of ANP on cAMP inhibition of thrombin-induced permeability (Figure 3B). Similar results were seen with EHNA (erythro-9-[2-hydroxy-3-nonyl]-adenine), a structurally different PDE2 inhibitor (supplemental Figure VI). This indicates that as cGMP concentrations increase its ability to potentiate the effects of cAMP by inhibiting PDE3 at lower concentrations is reversed by activation of PDE2.
cGMP Has Differential Effects on PDE2A and PDE3A cAMP Hydrolytic Activity
The functional studies of cGMP on thrombin-induced permeability suggest that cGMP biphasically regulates cAMP levels in HUVECs via inhibition of PDE3 at low concentrations and activation of PDE2 at higher concentrations. To demonstrate the effect of cGMP on PDE3 and PDE2 cAMP hydrolytic activity, we immunoprecipitated PDE2A and PDE3A from HUVECs and assayed for cAMP PDE activity at various concentrations of cGMP. As previously shown for PDE3A found in the heart and platelets, cGMP dose-dependently inhibited PDE3A activity with a calculated IC50 of 59 nmol/L (Figure 4). PDE2A activity, on the other hand, was stimulated nearly 30-fold by cGMP with a calculated EC50 of 1.1 µmol/L, which is similar to results found with PDE2A from the heart and adrenal gland (Figure 4). These results demonstrate that at low, nanomolar concentrations, cGMP inhibits PDE3A, whereas at higher, micromolar concentrations, cGMP activates PDE2A. Because the Vmax for PDE2A is much higher than that for PDE3A, one would expect that at high concentrations of cAMP, the PDE2A effect should dominate if both PDEs control pools of cAMP that can regulate endothelial barrier function.
|
cGMP Regulates cAMP in HUVECs via Inhibition of PDE3 and Activation of PDE2
These functional studies of cGMP on cAMP inhibition of thrombin-induced permeability suggested that cGMP can differentially regulate cAMP levels in HUVECs via PDE2 and PDE3. It is, however, not clear whether these effects could be detected on total cellular cAMP. When tested directly, we found little effect on whole cell cAMP (data not shown). However, HUVECs have been shown to express high levels of PDE4 in addition to PDE2 and PDE3.13 Thus, we reasoned that the pool(s) served by PDE2 and PDE3 might be more easily detected in cells in which PDE4 had been inhibited. Therefore, we first pretreated HUVECs with rolipram (10 µmol/L), a PDE4-specific inhibitor, and used the same conditions and stimuli described for in studying permeability. Under these conditions, we show that MPB–forskolin (0.1 µmol/L) causes a small but significant increase in cAMP after 5 minutes (supplemental Figure VIIA). MPB–forskolin plus inhibition of PDE3 by trequinsin or PDE2 by Bay 60-7550 increased cAMP in HUVECs compared with MPB–forskolin alone (supplemental Figure VIIA). Inhibition of both PDE2 and PDE3 had an additive effect on MPB–forskolin–stimulated cAMP increases. These results suggest that both PDE2 and PDE3 can regulate cAMP in HUVECs. When PDE3 was inhibited by trequinsin, HUVECs stimulated with MPB–forskolin and increasing amounts of ANP displayed a decrease in cAMP levels, suggesting cGMP activation of PDE2 (supplemental Figure VIIB). On the other hand, in the presence of Bay 60-7550, which inhibits PDE2, ANP caused an increase in cAMP, suggesting cGMP inhibition of PDE3 (supplemental Figure VIIC). Together, these results strongly suggest that cGMP can regulate cAMP levels in HUVECs via inhibition of PDE3 and activation of PDE2.
Alterations in the Expression and Activity of PDE2 and PDE3 in Endothelial Cells Modifies cGMP Effects on Endothelial Cell Permeability
Recently, it was shown that TNF-, an inflammatory cytokine known to increase endothelial cell permeability, also increased the expression of PDE2 in HUVECs.9 We confirmed an 7-fold increase in PDE2A mRNA and protein expression in HUVECs after 24 hours of TNF- stimulation (Figure 5A and 5B). Moreover, using quantitative real-time PCR, we found that PDE3A mRNA was decreased 50% in TNF-–stimulated HUVECs (Figure 5A). At the protein level, PDE3A expression decreased 25% in this time frame (Figure 5C). The apparent molecular mass of PDE3A expressed in HUVECs is 94 kDa, corresponding to PDE3A3.14 The recombinant PDE3A1 used as a positive control in Figure 5C has a molecular mass of 115 kDa. Importantly cAMP PDE activity levels also changed after TNF- stimulation. TNF- increased basal PDE2 activity from 16 pmol cAMP/min · mg–1 to nearly 100 pmol cAMP/min · mg–1 (Figure 5D). PDE3 activity, 60 pmol cAMP/min · mg–1, did not change significantly. Therefore, we used TNF- treatment as a means to alter the expression and activity of PDE2A and PDE3A. As shown previously, 0.3 nmol/L ANP potentiated the inhibitory actions of cAMP on thrombin-induced permeability (Figure 6). In contrast, in HUVECs pretreated with 10 ng/mL TNF- for 24 hours to induce PDE2 and decrease PDE3A expression, the same concentration of ANP was less effective in potentiating the inhibitory effects of cAMP (Figure 6). The PDE2 inhibitor Bay 60-7550 blocked this change in response to ANP (Figure 6). Again, these results are consistent with the hypothesis that agents able to increase the expression of PDE2 and decrease PDE3A expression can alter the response of endothelial cells to cGMP.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the current study, we have demonstrated that cGMP-elevating agents such as ANP and NO have biphasic effects on cAMP inhibition of thrombin-induced endothelial permeability. Furthermore, we show that a major reason for the biphasic actions of cGMP on cAMP inhibition of thrombin-induced endothelial permeability is attributable to a dose-dependent differential regulation of PDE2 and PDE3. Finally, we demonstrate that altering the expression of PDE2 and PDE3 in endothelial cells can alter the effect of cGMP on endothelial permeability (see model in Figure 7). These results likely explain many of the conflicting results reported in the literature concerning the actions of cGMP on endothelial permeability and suggest important roles for PDE2 and PDE3 in regulating endothelial function.
|
Whereas the observation that cAMP produced by transmembrane adenylate cyclases decreases endothelial permeability is well accepted, the role of cGMP in regulating endothelial permeability has been controversial. There are several reports in which cGMP has been implicated in increasing endothelial permeability, although just as many demonstrate a barrier-enhancing effect of cGMP.2,5,6,15 For example, mice injected with high doses of ANP have been shown to have increased endothelial permeability, whereas others have shown that ANP decreases thrombin-induced permeability.5,16 The use of NO synthase inhibitors, endothelial NO synthase knockout mice, and our own results with soluble guanylyl cyclase inhibitors suggest that at basal levels, NO-derived cGMP inhibits increases in endothelial permeability.6,17 Conversely others have shown that endothelial NO synthase is required for inflammatory increases in permeability.18 The present results demonstrating a biphasic effect of cGMP caused by differential regulation of cAMP hydrolyzing PDE2 and PDE3 may help reconcile these differences.
Our results demonstrating the differential effects of cGMP on PDE2A and PDE3A cAMP hydrolytic activity are in agreement with previous studies and, to our knowledge, are the first to measure the effect of cGMP in a dose-dependent manner on PDE2 and PDE3 activity isolated from the same cell type (Figure 4).19,20 Our findings that cGMP inhibits PDE3A with an IC50 nearly 20-fold lower than the EC50 for PDE2A activation supports our model of biphasic regulation of endothelial permeability by cGMP. It has been shown previously that small increases in cGMP can modulate cAMP-dependent processes, such as platelet aggregation via inhibition of PDE3.7 Furthermore, it has been shown that ANP, via increasing cGMP, can inhibit adrenocorticotropic hormone-induced aldosterone production from adrenal cells via activation of PDE2.7 Despite these reports, in cells expressing both PDE2 and PDE3, there have been few examples in which the ability of cGMP to differentially regulate cAMP signaling via these PDEs has been shown. One exception is cardiomyocytes, in which both PDE2 and PDE3 are expressed. Vandecasteele et al demonstrated that cGMP at low concentrations inhibited PDE3, resulting in increased PKA stimulation of L-type Ca2+ current.21 However, higher concentrations of cGMP, via activation of PDE2, attenuated this stimulation. These results are consistent with our findings in endothelial cells that low concentrations of cGMP inhibit PDE3 to enhance barrier function, whereas higher concentrations activate PDE2 to increase permeability. Our results are, to our knowledge, the first to demonstrate opposing roles for PDE2 and PDE3 in the cGMP regulation of endothelial permeability.
Part of the reason for the lack of consistent information about the roles of PDEs 2 and 3 relates to a limitation of 1 of the major approaches used by many groups, namely the use of cGMP analogs.6,22 Although analogs such as 8-Br-cGMP and 8-pCPT-cGMP are useful in demonstrating the involvement of PKG, they are not useful in determining the possible contribution of cGMP-regulated PDEs because neither 8-Br-cGMP nor 8-pCPT-cGMP is an effective regulator of PDE2 or PDE3.11 Furthermore, several groups have reported effects of ANP and NO on permeability but did not see similar results with cGMP analogs.23,24 As a result, these findings were interpreted as being cGMP independent. The data presented in this report argue that the permeability effects seen with ANP and NO probably actually are cGMP dependent but work via regulation of PDE2 and PDE3 rather than PKG. We and others have found low levels of expression of PKG in HUVECs that can be activated by cGMP analogs (supplemental Figure V); yet these analogs have little or no effect on regulation of endothelial permeability (Figure 2).25,26 The finding that the actions of cGMP on endothelial permeability is attributable to its differential regulation of PDE2 and PDE3 further underscores the need to examine the role these PDEs play in cGMP signaling.
Until recently, it had been accepted that cAMP improves endothelial barrier function via its ability to activate PKA.4 Lately, another cAMP effector molecule, Epac, has also been shown to mediate at least part of the barrier-enhancing actions of cAMP.10 Furthermore, it was recently reported that the effect of cAMP on endothelial permeability is dependent on where in the cell cAMP is produced.27 In general, the concept of compartmentalization of cyclic nucleotide signaling is becoming well accepted.28 It is possible that there are several cAMP "pools" that are responsible for the regulation of endothelial permeability via PKA and/or Epac and controlled by specific PDEs. The role of cGMP, PDE2, and PDE3 in the regulation of PKA, Epac, and/or different cAMP pools in the regulation of endothelial barrier function requires further investigation.
Normal physiological levels of ANP and NO are relatively low but increase significantly under pathological conditions.29,30 Our findings that picomolar concentrations of ANP enhance endothelial barrier function via inhibition of PDE3 are consistent with normal vascular function in healthy individuals. At higher ANP concentrations, as seen in heart failure, endothelial barrier function is compromised.5,29 Our results implicating activation of PDE2 by ANP fits well with those reported by Sabrane et al, in which mice lacking endothelial guanylyl-cyclase-A receptor for ANP had decreased permeability and exhibited hypertension and cardiac hypertrophy.15 Thus PDE2 regulation of endothelial permeability may play an important role in the physiological response to pathological conditions. NO has been traditionally thought of as an atheroprotective signal, and dysfunctional endothelium is reported to have decreased levels of NO.31,32 From the present results, low NO would be expected to lead to decreased inhibition of PDE3 by cGMP and thus increased permeability. At the same time, in pathological conditions such as acute inflammation, production of large amounts of NO by endothelial NO synthase and immune cells increases endothelial permeability in vitro and in vivo.18,33 The data reported here strongly suggest that this increased permeability is attributable to increased activity of PDE2. For example, TNF- and other inflammatory cytokines stimulate several endothelial cell responses that lead to endothelial dysfunction and pathology.1 The upregulation of PDE2 and downregulation of PDE3A by TNF- suggests that altering the expression and activity of PDE2 and PDE3, thus altering the effect of cGMP on endothelial cell barrier function, may be a major mechanism by which TNF- causes endothelial dysfunction (Figure 5). In particular, the finding that the effect of cGMP on endothelial permeability is altered in HUVECs stimulated with TNF- and that this change can be reversed by PDE2 inhibition supports this hypothesis (Figure 6). Thus PDE2 and PDE3 in the endothelium can act as a sensor or switch to detect normal versus pathological concentrations of cGMP and thus regulate endothelial permeability accordingly. Furthermore, the expression and activity levels of these PDEs control the sensitivity of this switch to cGMP.
|
Acknowledgments |
|---|
We thank Susan Chacona for help in obtaining umbilical cords, Valeria Vasta for designing primers for real-time PCR, and Xiao-Bo Tang for recombinant PDE3A1.
Sources of Funding
This work was supported by NIH grants DK021723 (to J.A.B.) and HL077789 (to C.Y.) and American Heart Association Research Grant 0455847T (to C.Y.).
Disclosures
None.
|
Footnotes |
|---|
Original received January 15, 2007; resubmission received April 16, 2007; revised resubmission received August 1, 2007; accepted August 7, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- Ross R. Atherosclerosis–an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
[Free Full Text] - van Nieuw Amerongen GP, van Hinsbergh VW. Targets for pharmacological intervention of endothelial hyperpermeability and barrier function. Vascul Pharmacol. 2002; 39: 257–272.[CrossRef][Medline] [Order article via Infotrieve]
- van Hinsbergh WM. Endothelial permeability for macromolecules. Mechanistic aspects of pathophysiological modulation. Arterioscler Thromb Vasc Biol. 1997; 17: 1018–1023.
[Free Full Text] - Mehta D, Malik AB. Signaling mechanisms regulating endothelial permeability. Physiol Rev. 2006; 86: 279–367.
[Abstract/Free Full Text] - Zimmerman RS, Trippodo NC, MacPhee AA, Martinez AJ, Barbee RW. High-dose atrial natriuretic factor enhances albumin escape from the systemic but not the pulmonary circulation. Circ Res. 1990; 67: 461–468.
[Abstract/Free Full Text] - Draijer R, Atsma DE, van der Laarse A, van Hinsbergh VW. cGMP and nitric oxide modulate thrombin-induced endothelial permeability. Regulation via different pathways in human aortic and umbilical vein endothelial cells. Circ Res. 1995; 76: 199–208.
[Abstract/Free Full Text] - Bender AT, Beavo JA. Cyclic nucleotide phosphodiesterases: molecular regulation to clinical use. Pharmacol Rev. 2006; 58: 488–520.
[Abstract/Free Full Text] - Netherton SJ, Maurice DH. Vascular endothelial cell cyclic nucleotide phosphodiesterases and regulated cell migration: implications in angiogenesis. Mol Pharmacol. 2005; 67: 263–272.
[Abstract/Free Full Text] - Seybold J, Thomas D, Witzenrath M, Boral S, Hocke AC, Burger A, Hatzelmann A, Tenor H, Schudt C, Krull M, Schutte H, Hippenstiel S, Suttorp N. Tumor necrosis factor-alpha-dependent expression of phosphodiesterase 2: role in endothelial hyperpermeability. Blood. 2005; 105: 3569–3576.
[Abstract/Free Full Text] - Fukuhara S, Sakurai A, Sano H, Yamagishi A, Somekawa S, Takakura N, Saito Y, Kangawa K, Mochizuki N. Cyclic AMP potentiates vascular endothelial cadherin-mediated cell-cell contact to enhance endothelial barrier function through an Epac-Rap1 signaling pathway. Mol Cell Biol. 2005; 25: 136–146.
[Abstract/Free Full Text] - Beltman J, Becker DE, Butt E, Jensen GS, Rybalkin SD, Jastorff B, Beavo JA. Characterization of cyclic nucleotide phosphodiesterases with cyclic GMP analogs: topology of the catalytic domains. Mol Pharmacol. 1995; 47: 330–339.[Abstract]
- Draijer R, Vaandrager AB, Nolte C, de Jonge HR, Walter U, van Hinsbergh VW. Expression of cGMP-dependent protein kinase I and phosphorylation of its substrate, vasodilator-stimulated phosphoprotein, in human endothelial cells of different origin. Circ Res. 1995; 77: 897–905.
[Abstract/Free Full Text] - Favot L, Keravis T, Holl V, Le Bec A, Lugnier C. VEGF-induced HUVEC migration and proliferation are decreased by PDE2 and PDE4 inhibitors. Thromb Haemost. 2003; 90: 334–343.[Medline] [Order article via Infotrieve]
- Yan C, Miller CL, Abe J. Regulation of phosphodiesterase 3 and inducible cAMP early repressor in the heart. Circ Res. 2007; 100: 489–501.
[Abstract/Free Full Text] - Sabrane K, Kruse MN, Fabritz L, Zetsche B, Mitko D, Skryabin BV, Zwiener M, Baba HA, Yanagisawa M, Kuhn M. Vascular endothelium is critically involved in the hypotensive and hypovolemic actions of atrial natriuretic peptide. J Clin Invest. 2005; 115: 1666–1674.[CrossRef][Medline] [Order article via Infotrieve]
- Baron DA, Lofton CE, Newman WH, Currie MG. Atriopeptin inhibition of thrombin-mediated changes in the morphology and permeability of endothelial monolayers. Proc Natl Acad Sci U S A. 1989; 86: 3394–3398.
[Abstract/Free Full Text] - Predescu D, Predescu S, Shimizu J, Miyawaki-Shimizu K, Malik AB. Constitutive eNOS-derived nitric oxide is a determinant of endothelial junctional integrity. Am J Physiol Lung Cell Mol Physiol. 2005; 289: L371–L381.
[Abstract/Free Full Text] - Hatakeyama T, Pappas PJ, Hobson RW 2nd, Boric MP, Sessa WC, Duran WN. Endothelial nitric oxide synthase regulates microvascular hyperpermeability in vivo. J Physiol. 2006; 574: 275–281.
[Abstract/Free Full Text] - Leroy MJ, Degerman E, Taira M, Murata T, Wang LH, Movsesian MA, Meacci E, Manganiello VC. Characterization of two recombinant PDE3 (cGMP-inhibited cyclic nucleotide phosphodiesterase) isoforms, RcGIP1 and HcGIP2, expressed in NIH 3006 murine fibroblasts and Sf9 insect cells. Biochemistry. 1996; 35: 10194–10202.[CrossRef][Medline] [Order article via Infotrieve]
- Martinez SE. PDE2 Structure and Function. In: Beavo JA, Francis SH, Houslay MD, eds. Cyclic Nucleotide Phosphodiesterases in Health and Disease. Boca Raton, Fla: CRC Press; 2007: 55–77.
- Vandecasteele G, Verde I, Rucker-Martin C, Donzeau-Gouge P, Fischmeister R. Cyclic GMP regulation of the L-type Ca(2+) channel current in human atrial myocytes. J Physiol. 2001; 533: 329–340.
[Abstract/Free Full Text] - Holschermann H, Noll T, Hempel A, Piper HM. Dual role of cGMP in modulation of macromolecule permeability of aortic endothelial cells. Am J Physiol. 1997; 272: H91–H98.[Medline] [Order article via Infotrieve]
- Gupta MP, Ober MD, Patterson C, Al-Hassani M, Natarajan V, Hart CM. Nitric oxide attenuates H(2)O(2)-induced endothelial barrier dysfunction: mechanisms of protection. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L116–L126.
[Abstract/Free Full Text] - Klinger JR, Warburton R, Carino GP, Murray J, Murphy C, Napier M, Harrington EO. Natriuretic peptides differentially attenuate thrombin-induced barrier dysfunction in pulmonary microvascular endothelial cells. Exp Cell Res. 2006; 312: 401–410.[Medline] [Order article via Infotrieve]
- Hood J, Granger HJ. Protein kinase G mediates vascular endothelial growth factor-induced Raf-1 activation and proliferation in human endothelial cells. J Biol Chem. 1998; 273: 23504–23508.
[Abstract/Free Full Text] - Rivero-Vilches FJ, de Frutos S, Saura M, Rodriguez-Puyol D, Rodriguez-Puyol M. Differential relaxing responses to particulate or soluble guanylyl cyclase activation on endothelial cells: a mechanism dependent on PKG-I alpha activation by NO/cGMP. Am J Physiol Cell Physiol. 2003; 285: C891–C898.
[Abstract/Free Full Text] - Sayner SL, Alexeyev M, Dessauer CW, Stevens T. Soluble adenylyl cyclase reveals the significance of cAMP compartmentation on pulmonary microvascular endothelial cell barrier. Circ Res. 2006; 98: 675–681.
[Abstract/Free Full Text] - Fischmeister R, Castro LR, Abi-Gerges A, Rochais F, Jurevicius J, Leroy J, Vandecasteele G. Compartmentation of cyclic nucleotide signaling in the heart: the role of cyclic nucleotide phosphodiesterases. Circ Res. 2006; 99: 816–828.
[Abstract/Free Full Text] - Potter LR, Abbey-Hosch S, Dickey DM. Natriuretic peptides, their receptors, and cyclic guanosine monophosphate-dependent signaling functions. Endocr Rev. 2006; 27: 47–72.
[Abstract/Free Full Text] - Salerno L, Sorrenti V, Di Giacomo C, Romeo G, Siracusa MA. Progress in the development of selective nitric oxide synthase (NOS) inhibitors. Curr Pharm Des. 2002; 8: 177–200.[CrossRef][Medline] [Order article via Infotrieve]
- Endemann DH, Schiffrin EL. Endothelial dysfunction. J Am Soc Nephrol. 2004; 15: 1983–1992.
[Abstract/Free Full Text] - Li H, Wallerath T, Munzel T, Forstermann U. Regulation of endothelial-type NO synthase expression in pathophysiology and in response to drugs. Nitric Oxide. 2002; 7: 149–164.[CrossRef][Medline] [Order article via Infotrieve]
- Bucci M, Roviezzo F, Posadas I, Yu J, Parente L, Sessa WC, Ignarro LJ, Cirino G. Endothelial nitric oxide synthase activation is critical for vascular leakage during acute inflammation in vivo. Proc Natl Acad Sci U S A. 2005; 102: 904–908.
[Abstract/Free Full Text]
This article has been cited by other articles:
 |
|
 |
|
  B. Schreier, S. Borner, K. Volker, S. Gambaryan, S. C. Schafer, P. Kuhlencordt, B. Gassner, and M. Kuhn The Heart Communicates with the Endothelium through the Guanylyl Cyclase-A Receptor: Acute Handling of Intravascular Volume in Response to Volume Expansion Endocrinology, August 1, 2008; 149(8): 4193 - 4199. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||