This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 91/7/618    most recent 01.RES.0000036603.61868.F9v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hippenstiel, S. Articles by Suttorp, N. Search for Related Content PubMed PubMed Citation Articles by Hippenstiel, S. Articles by Suttorp, N. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB CALCIUM COMPOUNDS CALCIUM, ELEMENTAL HYDROGEN PEROXIDE Related Collections Cell biology/structural biology Pulmonary biology and circulation Pulmonary circulation and disease Other Vascular biology

(Circulation Research. 2002;91:618.)
© 2002 American Heart Association, Inc.

Integrative Physiology

Adrenomedullin Reduces Endothelial Hyperpermeability

Stefan Hippenstiel, Martin Witzenrath, Bernd Schmeck, Andreas Hocke, Mathias Krisp, Matthias Krüll, Joachim Seybold, Werner Seeger, Wolfgang Rascher, Hartwig Schütte, Norbert Suttorp

From Charité, Department of Internal Medicine (S.H., M.W., B.S., A.H., M. Krisp, M. Krüll, J.S., N.S.), Humboldt-University, Berlin, Germany; the Department of Pediatrics (W.R.), University of Erlangen-Nürnberg, Erlangen, Germany; and the Department of Internal Medicine (W.S., H.S.), Justus-Liebig University, Giessen, Germany.

Correspondence to Stefan Hippenstiel, MD, Charité, Department of Internal Medicine/Infectious Diseases, Humboldt-University, Augustenburger Platz 1, 13353 Berlin, Germany. E-mail stefan.hippenstiel{at}charite.de

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Endothelial hyperpermeability induced by inflammatory mediators is a hallmark of sepsis and adult respiratory distress syndrome. Increased levels of the regulatory peptide adrenomedullin (ADM) have been found in patients with systemic inflammatory response. We analyzed the effect of ADM on the permeability of cultured human umbilical vein endothelial cell (HUVEC) and porcine pulmonary artery endothelial cell monolayers. ADM dose-dependently reduced endothelial hyperpermeability induced by hydrogen peroxide (H₂O₂), thrombin, and Escherichia coli hemolysin. Moreover, ADM pretreatment blocked H₂O₂-related edema formation in isolated perfused rabbit lungs and increased cAMP levels in lung perfusate. ADM bound specifically to HUVECs and porcine pulmonary artery endothelial cells and increased cellular cAMP levels. Simultaneous inhibition of cAMP-degrading phosphodiesterase isoenzymes 3 and 4 potentiated ADM-dependent cAMP accumulation and synergistically enhanced ADM-dependent reduction of thrombin-induced hyperpermeability. However, ADM showed no effect on endothelial cGMP content, basal intracellular Ca²⁺ levels, or the H₂O₂-stimulated, thrombin-stimulated, or Escherichia coli hemolysin–stimulated Ca²⁺ increase. ADM diminished thrombin- and H₂O₂-related myosin light chain phosphorylation as well as stimulus-dependent stress fiber formation and gap formation in HUVECs, suggesting that ADM may stabilize the barrier function by cAMP-dependent relaxation of the microfilament system. These findings identify a new function of ADM and point to ADM as a potential interventional agent for the reduction of vascular leakage in sepsis and adult respiratory distress syndrome.

Key Words: adrenomedullin • cultured endothelial cells • endothelial permeability • endothelial barrier dysfunction

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The incidence of sepsis and ensuing multiple organ failure has increased over the past two decades and has caused multiple deaths in intensive care units. The development of adult respiratory distress syndrome (ARDS) characterized by noncardiogenic pulmonary edema contributes substantially to a fatal outcome.¹ Increased microvascular permeability is a hallmark of an inflammatory reaction, including ARDS-related pulmonary edema formation. Circumstantial evidence has suggested that endothelial hyperpermeability is related to alterations of the cellular cytoskeleton.² Endothelial cells have been shown to contain an elaborate microfilament system allowing active actin- and myosin-based cell contraction. Activation of cell contraction and disturbance of junctional organization subsequently result in the induction of interendothelial gaps followed by enhanced paracellular endothelial permeability.^2–6 Major initiators of this process are polymorphonuclear leukocyte–derived oxygen metabolites,^7–9 pore-forming bacterial exotoxins,^9,10 and endogenous proinflammatory mediators, such as thrombin.^5,9,11

Exposure of endothelial cells to hydrogen peroxide (H₂O₂) results in the activation of multiple signaling pathways, finally leading to the activation of myosin light chain (MLC) kinase and disrupted barrier function.^7,8 Thrombin-activated Rho kinase increases the phosphorylation of endothelial cell MLC by directly phosphorylating MLC and simultaneously inhibiting MLC phosphatase, thereby maximally enhancing MLC phosphorylation.^4,5 The hemolysin of Escherichia coli (HlyA) is an important prototype of pore-forming bacterial exotoxins that allows Ca²⁺ influx into the cytosol of target cells.^9,12,13 All three stimuli (H₂O₂, thrombin, and HlyA) have been shown to induce endothelial cell contraction followed by endothelial hyperpermeability in vitro and in vivo.

Previous studies from others and our group have demonstrated an improvement of endothelial barrier function by the activation of adenylyl cyclase and/or phosphodiesterase (PDE) inhibition.^2,8,9,14 Analysis of the endothelial cell PDE isoenzyme pattern showed high activities of PDE2 to PDE4. PDE2 mainly metabolizes cGMP, whereas PDE3/4 is specific for cAMP.^8,9

Accumulating evidence suggests a pivotal role of adrenomedullin (ADM), a multifunctional regulatory peptide, in sepsis and septic shock.^15–19 ADM, a 52-amino-acid peptide belonging to the calcitonin gene–related peptide family, participates in the control of central body functions, such as vascular tone regulation or fluid and electrolyte homeostasis (see reviews^20,21). In systemic inflammatory response, plasma levels of ADM in vertebrates, including human beings, were found to be elevated.^15,17–19 Endothelial cells produce ADM in response to proinflammatory cytokines or lipopolysaccaride²² under the control of the transcription factors nuclear factor for interleukin-6 and activator protein 2.²²

Calcitonin receptor–like receptor (CRLR) and receptor activity–modifying protein (RAMP)2 and RAMP3 together form ADM receptors coupled to cholera toxin–sensitive G proteins; however, alternative ADM binding receptors may exist.^20,23 Stimulation of virtually all cells currently investigated with ADM has resulted in a marked increase in cellular cAMP content and subsequent activation of protein kinase A (PKA).^20,21,24 In contrast, conflicting results regarding an ADM-induced increase in [Ca²⁺]_i have been reported.^24,25

Because transgenic mice overexpressing ADM in their vasculature have turned out to be resistant to lipopolysaccharide-induced shock, the role of ADM in sepsis deserves special consideration.²⁶ Moreover, mice with disrupted ADM genes have displayed an extreme hydrops fetalis and cardiovascular abnormalities, suggesting a central role of ADM in the regulation of the cardiovascular system, especially endothelial cell function.²⁷

In the present investigation, we tested the hypothesis that elevated ADM levels in systemic inflammatory response stabilize endothelial barrier function by preventing endothelial cell contraction and paracellular fluid flux. ADM treatment diminished endothelial hyperpermeability induced by stimuli as diverse as oxygen metabolites, pore-forming bacterial exotoxins, and endogenous proinflammatory mediators. Moreover, ADM blocked H₂O₂-related lung edema formation in a model of isolated rabbit lungs. ADM induced cAMP formation, reduced thrombin- and H₂O₂-induced MLC phosphorylation, and prevented endothelial cell contraction. Overall, our data suggest that ADM may act as a counterregulatory peptide in systemic inflammatory response by improvement of endothelial barrier function.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Preparation of Endothelial Cells
Human umbilical vein endothelial cells (HUVECs) and porcine pulmonary artery endothelial cells (PAECs) were prepared as previously described.^8,9,13,28,29

Radioligand Binding for ADM
Competitive receptor binding studies for human ADM were performed with minor modifications as described earlier for neuropeptide Y.³⁰

Determination of HUVEC Cyclic Nucleotide Content
Cyclic nucleotide content was measured by using a commercially available ELISA (Biotrend).

Analysis of Endothelial Permeability
Hydraulic conductivity of endothelial cell monolayers was determined as described previously.^8,9,31 Briefly, a confluent cell monolayer on a filter membrane was mounted in a modified chemotaxis chamber, and a hydrostatic pressure of 10² mm H₂O was applied to the "luminal" side of the cell monolayer. The filtration rate across the endothelial monolayer was continuously determined, and the hydraulic conductivity was calculated and expressed as 10^-5 cm · s^-1 · cm H₂O^-1.

F-Actin Staining
Cells were fixed in paraformaldehyde and permeabilized, and actin was stained by using phalloidin Alexa 488 (Molecular Probes). Cells were analyzed with the use of a Pascal 5 confocal scanning laser microscope (Zeiss).

Detection of MLC
Cells were harvested in lysis buffer containing phosphatase and protease inhibitors. Equal amounts of lysates were subjected to SDS-PAGE (12.5% gel) and blotted. Membranes were simultaneously exposed to goat phospho-specific MLC (Thr18/Ser19) and rabbit extracellular signal–regulated kinase (ERK)1-specific antibody (both Santa Cruz Biotechnology) incubated with secondary antibodies (IRDye 800–labeled anti-goat and Cy5.5-labeled anti-rabbit, respectively). Proteins were detected by using an Odyssey infrared imaging system (LI-COR Inc).

Determination of [Ca²⁺]_i
HUVECs cultured on glass coverslips were loaded with the fluorescent Ca²⁺-sensitive dye fura 2 and analyzed by use of a fluorescence spectrophotometer (Instruments S.A.). Excitation wavelength was alternated between 343 and 380 nm. Emitted light was detected at 510 nm. Fura 2 fluorescence was calibrated according to the method described by Grynkiewicz et al.³²

Lung Model
A model of perfused rabbit lungs has previously been described in detail (see overview³³). Briefly, rabbits were ventilated with room air by use of a Harvard respirator (Hugo Sachs Elektronik). Catheters were placed into the pulmonary artery and left atrium, and the lungs were perfused with Krebs-Henseleit buffer. In parallel, room air ventilation was supplemented with 5% CO₂. Lungs included in the present study had a homogeneous white appearance with no signs of hemostasis, edema, or atelectasis, and they had pulmonary artery and ventilation pressures in the normal range and were isogravimetric during an initial steady-state period of at least 30 minutes.

Capillary filtration coefficient (Kfc, normalized for wet lung weight) and total vascular compliance were determined gravimetrically, and lung weight gain was calculated. Lungs were perfused for 120 minutes in the absence or presence of 10^-7 mol/L ADM. Perfusate (100 µmol H₂O₂ · min^-1 · 150 mL^-1) was infused as indicated. In all experiments, 5 µmol/L of thromboxane receptor antagonist BM 13.505 was admixed with the recirculating buffer fluid 10 minutes before stimulus application. Perfusate samples for determination of cAMP were taken after ADM stimulation and were processed for ELISA (Biotrend).

Statistical Methods
Depending on the number of groups (A) and the number of different time points studied (B), data of Figures 1, 2, and 6 were analyzed by an AxB ANOVA. A one-way ANOVA was used for data of Figures 3 and 4A. Main effects were then compared by an F probability test. A value of P<0.05 was considered significant. Data are displayed as mean±SEM.

View larger version (30K): [in this window] [in a new window]   Figure 1. ADM dose-dependently reduced hyperpermeability of PAECs provoked by H2O2. Resting (sealed) PAEC monolayers displayed a hydraulic conductivity of <0.5x10-5 cm · s-1 · cm H2O-1. Pretreatment of sealed PAEC monolayers with 0.01 to 1 µmol/L ADM stabilized barrier function of H2O2 (10-3 mol/L)–exposed endothelial cells. Control monolayers were stable over the entire experimental period. Data are mean±SEM of 4 separate experiments. #P<0.05 vs H2O2.

View larger version (24K): [in this window] [in a new window]   Figure 3. ADM stimulated cAMP but not cGMP formation in HUVECs. Treatment of HUVECs for 10 minutes with increasing concentrations of ADM (0.1 nmol/L to 10 µmol/L) induced cAMP formation in these cells (A). Inhibition of PDE3/4 with 10 µmol/L dual-selective PDE inhibitor zardaverine (Zarda) for 30 minutes dramatically increased ADM (1 to 100 nmol/L)–dependent cAMP accumulation in HUVECs (A). ADM or ADM in combination with PDE3/4 inhibitor increased cAMP content to a lesser extent than 1 µmol/L forskolin or forskolin in combination with Zarda. Incubation of HUVECs with ADM (100 µmol/L) alone or PDE inhibition of cGMP-degrading PDE2 (10 µmol/L EHNA) 15 minutes before ADM addition did not alter cGMP content (B). In contrast, the NO donor sodium nitroprusside (SNP) alone or in combination with the PDE2 inhibitor increased cGMP content (B). Data are mean±SEM of 4 separate experiments. For panel A, *P<0.05 vs control; #P<0.05 vs ADM and vs Zarda; and P<0.05 vs Zarda. For panel B, *P<0.05 vs control; #P<0.05 vs EHNA.

View larger version (40K): [in this window] [in a new window]   Figure 2. ADM dose-dependently diminished hyperpermeability of HUVEC monolayers induced by H2O2 (A), HlyA (B), and thrombin (Thr) (C). Hydraulic conductivity of resting HUVECs was <0.5x10-5 cm · s-1 · cm H2O-1. Pretreatment of HUVEC monolayers with 0.1 to 1 µmol/L ADM stabilized barrier function of H2O2 (10-3 mol/L)–exposed endothelial cells (A). Control monolayers were stable and responded promptly upon addition of staphylococcal -toxin (*), an established permeabilizing agent (A). Dose-dependent reduction of HlyA-induced increase in endothelial cell monolayer permeability by ADM is shown (B): cells were preincubated for 15 minutes with 0.1 to 1 µmol/L ADM before 0.1 HU/mL HlyA was added. ADM dose-dependently reduced Thr-related endothelial hyperpermeability (C). Preexposure of HUVEC monolayers to 0.01 to 1 µmol/L ADM 15 minutes before Thr stimulation dose-dependently blocked Thr-induced endothelial barrier dysfunction. Subthreshold concentrations of ADM (1 nmol/L) or dual-selective PDE3/4 inhibitor zardaverine (Za, 10 µmol/L), which per se had no significant effect on Thr-related hyperpermeability, were very effective when added in combination (D), suggesting that cAMP elevation by ADM and PDE3/4 inhibition acted synergistically with respect to endothelial barrier integrity. Data presented in panels A and B are mean±SEM of 4 separate experiments. Results shown in panels C and D are mean±SEM of 5 separate experiments. #P<0.05 vs H2O2 in panel A; #P<0.05 vs HlyA in panel B; #P<0.05 vs Thr in panel C; and #P<0.05 vs H2O2 vs ADM vs Za in panel D.

View larger version (30K): [in this window] [in a new window]   Figure 4. ADM reduced Thr- and H2O2-related MLC phosphorylation in HUVECs. Preincubation of cells with 1 µmol/L ADM reduced Thr (0.1 U/mL)–related and H2O2 (1 mmol/L)–related MLC phosphorylation as shown by Western blot with use of MLC-phospho-Thr18/Ser19–specific antibody (A and B). Phosphorylated MLC (P-MLC) and ERK1 were simultaneously detected by an infrared imaging system. MLC phosphorylation was normalized to detected protein of ERK1; total amount is demonstrated in relation to Thr- or H2O2-induced MLC phosphorylation (A). Data presented in panel A are mean±SEM of 3 separate experiments. #P<0.05 vs Thr (A, left); #P<0.05 vs H2O2 (A, right). Representative gels out of 3 separate experiments are shown in panel B.

View larger version (16K): [in this window] [in a new window]   Figure 6. ADM blocked H2O2-related lung edema formation in perfused rabbit lungs. Bolus addition of 0.1 nmol/L ADM 15 minutes before application of H2O2 (10 mmol/L) reduced the H2O2-related increase in lung Kfc, suggesting a profound endothelial barrier stabilization in ADM-treated lungs (A). cAMP perfusate levels increased after ADM bolus addition (B). Exposure of lungs to 0.1 µmol/L ADM and 5 mmol/L H2O2 resulted in increased cAMP generation in lungs, whereas H2O2 addition alone showed no effect on cAMP formation. Data are mean±SEM of 5 separate experiments each. #P<0.05 vs ADM+H2O2.

An expanded Materials and Methods section can be found in an online data supplement available at http://www.circresaha.org.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
ADM Improves Barrier Function of Cultured Endothelial Cell Monolayers
Sealed PAEC monolayers (Figure 1) and HUVEC monolayers (Figure 2) displayed a hydraulic conductivity of <0.5x10⁵ cm · s^-1 · cm H₂O^-1. The addition of 1 µmol/L ADM as a bolus to PAEC (Figure 1) or HUVEC (Figure 2) monolayers and continuous infusion of 10 µmol/L ADM had no effect on the endothelial barrier function of resting cells within the time frame tested (90 minutes, data not shown).

First, we used H₂O₂ to mimic a polymorphonuclear leukocyte–mediated oxidant attack (Figures 1 and 2A). H₂O₂ (1 mmol/L) time-dependently increased the endothelial monolayer permeability of PAECs (Figure 1) and HUVECs (Figure 2A). Pretreatment of PAECs with 0.01 to 1 µmol/L ADM or of HUVEC with 0.1 to 1 µmol/L ADM 15 minutes before the stimulus dose-dependently reduced the H₂O₂-related increase in endothelial hydraulic conductivity (Figures 1 and 2A). Control monolayers were stable throughout the experimental period and responded promptly to the addition of staphylococcal -toxin, a well-established permeabilizing agent, as shown for HUVECs (Figure 2A).

Second, HlyA was used as a prototype for a pore-forming bacterial toxin (Figure 2B). HlyA (0.1 hemolytic units [HU]) given as a bolus induced a rapid and strong increase in the hydraulic conductivity of HUVEC monolayers. Incubation of HUVECs with 0.1 to 1 µmol/L ADM 15 minutes before the addition of HlyA reduced the toxin-related increase in permeability (Figure 2B).

Third, we analyzed the effect of ADM on thrombin-mediated endothelial hyperpermeability (Figure 2C). Exposure of endothelial cell monolayers to 0.1 U thrombin resulted in a loss of barrier function within 20 minutes. The addition of 0.01 to 1 µmol/L ADM before thrombin stimulation substantially reduced the thrombin-induced hyperpermeability (Figure 2C). We then examined whether the inhibition of cAMP-specific PDE3/4 in combination with cAMP-elevating ADM had an effect on thrombin-mediated hyperpermeability (Figure 2D). Exposures of endothelial cell monolayers to concentrations of ADM (1 nmol/L, 15 minutes) and the PDE 3/4 inhibitor zardaverine (10 µmol/L, 30 minutes), which alone had no significant effect on thrombin-related hyperpermeability, were very effective when used in combination (Figure 2D).

ADM Binds to Endothelial Cells and Induces Formation of cAMP but Does Not Change cGMP or [Ca²⁺]_i Levels
Human ¹²⁵I-ADM binds to confluent HUVEC monolayers with a B_max of 0.4 pmol per well and displays a K_d of 147.9 nmol/L. The specificity of ADM binding to HUVECs was confirmed by competition experiments using increasing concentrations of unlabeled ADM. Confluent PAEC cultures showed a B_max of 0.25 pmol per well and a K_d of 13.4 nmol/L for human ¹²⁵I-ADM, which could be blocked by increasing concentrations of unlabeled ADM, indicating specificity of peptide binding (see online data supplement).

We also confirmed the ability of ADM to stimulate cAMP formation in endothelial cells. Treatment of HUVECs or PAECs for 10 minutes with increasing concentrations of ADM (0.1 nmol/L to 10 µmol/L) induced cAMP formation in HUVECs (Figure 3A; data for PAECs are not shown) to a lesser extent than exposure to the positive control forskolin (1 µmol/L) for 10 minutes. Previous studies had shown high activities of PDE isoenzymes 2 to 4 in endothelial cells and a multiplying effect of specific PDE inhibition on cyclic nucleotide accumulation.^8,9 To determine the effect of PDE inhibition on ADM-stimulated cAMP accumulation, cells were pretreated with 10 µmol/L of the dual-selective PDE3/4 inhibitor zardaverine for 30 minutes, which resulted in a dramatic rise of ADM- or forskolin-mediated cAMP accumulation (Figure 3A). In contrast, cGMP accumulation in HUVECs (Figure 3B) or PAECs (data not shown) was unaffected after ADM stimulation (up to 100 µmol/L ADM for 5, 15, or 30 minutes) even in the presence of erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA) (10 µmol/L), a PDE2 inhibitor (Figure 3B; data for 5 and 30 minutes are not shown). However, in the same experimental setup, a 4-fold increase of cGMP could be demonstrated when combining the PDE2 inhibitor EHNA with the NO donor sodium nitroprusside (10 µmol/L, 15 minutes) (Figure 3B).

An increase in [Ca²⁺]_i induced by thrombin, H₂O₂, or HlyA treatment of endothelial cells was considered to be a strong signal for endothelial cell retraction followed by endothelial hyperpermeability. ADM exposure of HUVECs (1 to 100 µmol/L) had no effect on basal [Ca²⁺]_i content within an observation period of 15 minutes (data not shown). We tested the hypothesis that ADM may reduce the H₂O₂-, thrombin-, or HlyA-mediated rise in [Ca²⁺]_i, thereby preventing F-actin rearrangement and hyperpermeability. However, pretreatment of HUVECs with 10 µmol/L ADM before stimulation with H₂O₂, thrombin, or HlyA showed no effect on the [Ca²⁺]_i increase induced by these particular agents (see online data supplement).

ADM Reduces Stimulus-Induced Phosphorylation of MLCs and Alterations of Human Endothelial Cell Filamentous Actin
Phosphorylation of regulatory MLC contributes significantly to endothelial cell contraction provoked by proinflammatory agents, thereby allowing paracellular fluid flux. Using an antibody directed against phospho-Thr18/phospho-Ser19 of MLC, we demonstrated that ADM reduced thrombin- and H₂O₂-dependent MLC phosphorylation (Figures 4A and 4B). Simultaneous detection of ERK1 confirmed equal protein loading.

In addition, ADM blocked H₂O₂-, thrombin-, or HlyA-mediated alterations of endothelial filamentous actin (F-actin), as shown by fluorescence microscopy (Figure 5). HUVECs exposed to solvent (Figure 5A) or 1 µmol/L ADM (Figure 5B) displayed a well-organized peripheral dense band of F-actin with only a few stress fibers. After stimulation of endothelial cells with 1 mmol/L H₂O₂ for 30 minutes (Figure 5C), 1 U thrombin for 15 minutes (Figure 5E), or 0.1 HU HlyA for 15 minutes (Figure 5G), the amount and density of stress fibers increased, whereas the peripheral dense band was disrupted, and gaps between the endothelial cells were opened. In contrast, in cells pretreated with 1 µmol/L ADM 15 minutes before stimulation with these agents (Figures 5D, 5F, and 5H), F-actin distribution remained unchanged, and intercellular gaps were closed as in the control monolayers.

View larger version (106K): [in this window] [in a new window]   Figure 5. ADM prevented H2O2-, Thr-, and HlyA-dependent microfilament reorganization and cell retraction. HUVECs treated with solvent (A) or 0.1 µmol/L ADM (B) showed no stress fiber or gap formation as visualized by phalloidin Alexa 488. Preexposure of cells for 15 minutes to 0.1 µmol/L ADM blocked H2O2-related (D), Thr-related (F), and HlyA-related (H) stress fiber and gap formation. Representative fields of HUVEC monolayers (out of 5 for panels A, B, E, and F; out of 3 for panels C, D, G, and H) are shown.

Human ADM Stimulates cAMP Formation and Reduces H₂O₂-Induced Vascular Hyperpermeability in Isolated Rabbit Lungs
We made use of isolated, ventilated, blood-free perfused rabbit lungs stimulated with H₂O₂ (100 µmol/min admixed to the 150 mL perfusate over 15 minutes) to analyze the power of ADM in the regulation of endothelial barrier function in a more integrated model (Figure 6). In lungs exposed to H₂O₂ alone, Kfc increased up to 9.0±2.08 within 45 minutes, followed by massive pulmonary edema formation (Figure 6A). In contrast, ADM (0.1 µmol/L ADM 15 minutes before H₂O₂ admixture) almost completely prevented H₂O₂-induced edema formation (Figure 6A). The vascular compliance remained unchanged and was not different between the experimental groups. Pulmonary artery pressure showed some minor elevation on H₂O₂ infusion but did not display a significant difference between ADM-treated and -untreated lungs at the time points of the Kfc determination (see online data supplement). Moreover, only negligible changes (1 mm Hg) of microvascular pressures were observed in both experimental groups, indicating that the reduced edema formation observed in the ADM-stimulated rabbit lungs was not due to reduced filtration pressure (see online data supplement).

cAMP content in the perfusate collected from the rabbit lungs processed for analysis of Kfc was strongly increased in ADM-exposed lungs (Figure 6B).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Recent studies have demonstrated elevated plasma levels of ADM in vertebrates with a systemic inflammatory response.^15–19 However, the role of ADM in the complex and dynamic disease process of sepsis is still largely undefined. On the one hand, the high ADM plasma levels observed in septic humans may contribute to hypotension and hyperdynamic circulatory response in sepsis, thereby contributing to disease progress.¹⁸ On the other hand, transgenic mice overexpressing ADM in their vasculature turned out to be resistant against lipopolysaccharide-induced shock, suggesting a rather beneficial effect of elevated ADM levels in sepsis.²⁶

Considering that endothelial hyperpermeability is the hallmark of an inflammatory reaction^1,2 and that mice lacking a functional ADM gene displayed an extreme hydrops fetalis,²⁷ we tested the hypothesis that elevated ADM levels stabilized endothelial barrier function, thereby acting as a "protective" peptide in the systemic inflammatory response. Our results clearly indicate that ADM potently stabilized endothelial barrier function. ADM preincubation reduced endothelial hyperpermeability induced by thrombin, H₂O₂, or HlyA in vitro. Moreover, in isolated perfused rabbit lungs, ADM decreased H₂O₂-related capillary hyperpermeability and edema formation. Specific binding of ADM to endothelial cells elevated intracellular cAMP and reduced MLC phosphorylation, thereby stabilizing the endothelial cell microfilament system.

Because a broad variety of stimuli may contribute to endothelial barrier dysfunction in sepsis, we chose three typical permeability-inducing agents to investigate the effects of ADM. H₂O₂, released by polymorphonuclear granulocytes in inflammatory reactions, activates complex signaling pathways in endothelial cells, including active myosin-based cell contraction² and protein kinase C (PKC) activation.⁷ HlyA, as a prototype of a bacterial pore-forming exotoxin, allows Ca²⁺ influx into the cytosol according to the transmembrane Ca²⁺ gradient and potently induces NO production¹³ as well as the expression of endothelial cell adhesion molecules.²⁹ Thrombin exposure of endothelial cells activates phospholipid hydrolysis, increases [Ca²⁺]_i, and promotes cell contraction.^2,5,11

All three stimuli induced active actin-myosin–based cell contraction, thereby increasing enhanced paracellular endothelial permeability. Notably, ADM preexposure of endothelial cell monolayers greatly reduced the increase in hydraulic conductivity in response to all three agents. It has been suggested that ADM-mediated cAMP elevation contributes to ADM effects in the vasculature.^19,24,25 In line with previous studies,^20,21,24,25 ADM incubation increased cAMP content in cultured endothelial cells. Human and porcine endothelial cells contain PDE3 and PDE4 for the degradation of cAMP⁸ and PDE2 for the degradation of cGMP.⁹ Inhibition of PDE3/4 with the dual-selective PDE3/4 inhibitor zardaverine significantly increased ADM-related cAMP accumulation and strengthened the endothelial barrier–protective effect of ADM as assessed for thrombin-related hyperpermeability. ADM seems to be as potent as a pharmacological stimulator of cAMP elevation, regarding maintenance of endothelial barrier function.^8,9 Moreover, ADM treatment elevated cAMP levels in isolated perfused rabbit lungs and reduced H₂O₂-related edema formation. This is consistent with previous studies showing that cAMP elevation potently blocks hyperpermeability in isolated rabbit lungs.³⁴

To exclude the effects of H₂O₂-dependent pulmonary vasoconstriction on edema formation, we used the thromboxane receptor antagonist BM 13.505. Although ADM was known to reduce systemic blood pressure^15,35 and, in some systems, pulmonary hypertension,³⁶ no substantial changes in pulmonary artery perfusion pressure were noted in H₂O₂- or solvent-exposed rabbit lungs pretreated with BM 13.505.

In line with the presently described role of ADM in the regulation of endothelial permeability under inflammatory conditions, recent observations in mice lacking a functional ADM gene point to a general role of ADM in permeability regulation: mice lacking a functional ADM gene displayed an extreme hydrops fetalis as well as cardiovascular abnormalities and died at mid gestation.²⁷ Overall, these observations suggest a pivotal role of ADM in the regulation of endothelial barrier function.

Alterations of the endothelial cell microfilament system are accompanied by active actin-myosin–based cell contraction, allowing increased paracellular fluid flux, which seems to be critical for edema formation under inflammatory conditions.^{1,2,4,5,8,9,11} ADM reduced thrombin- and H₂O₂-related phosphorylation of MLC and blocked endothelial cell contraction, intercellular gap formation, and stress fiber formation. Besides classic Ca²⁺/calmodulin-dependent MLC kinases, ³ Rho kinase may contribute to thrombin-related MLC phosphorylation. Although MLC kinase phosphorylates MLC at Ser19 and Thr18 as a sole mode of action, Rho kinase additionally blocks myosin phosphatase type 1, thereby enhancing MLC phosphorylation.^4,5,37 Inasmuch as cAMP elevation reduced Rho kinase–dependent phosphorylation of MLC in lipopolysaccharide-exposed endothelial cells³⁸ and Rho kinase was identified as a central regulator of thrombin-induced endothelial cell contraction,^4,5,37 ADM-related increased cAMP may act via inhibition of the Rho–Rho kinase pathway. Moreover, PKC-dependent phosphorylations of MLC and important permeability-regulating junctional proteins, such as vasodilator-stimulated phosphoprotein,³⁹ zonula occludens protein-1,^2,39 and vascular endothelial cadherin,² significantly contribute to barrier dysfunction. Although cAMP elevation seems not to prevent stimulus-dependent PKC activation in endothelial cells,⁴⁰ cAMP-related PKA activation may counterregulate PKC-induced phosphorylation effects. For example, it has been shown that PKA-dependent phosphorylation of the Ser157 residue of vasodilator-stimulated phosphoprotein diminishes paracellular permeability through the relaxation of actin cytoskeletal tension.³⁹

The data presented support the notion that ADM binds via specific receptors to endothelial cells. However, the situation is complex inasmuch as CRLR and RAMP2 and RAMP3 together form ADM receptors.^20,21,23 Moreover, alternative ADM binding sites may exist. Interestingly, the expression of CRLR-RAMP2/CRLR-RAMP3 complexes apparently undergoes regulatory changes in different tissues and stages during sepsis, thereby contributing to ADM-related effects in systemic inflammation.⁴¹

Besides the widely accepted ADM-dependent cAMP elevation, conflicting results were reported regarding an ADM-induced rise in [Ca²⁺]_i in endothelial cells.^20,24,25 In the present study, ADM exposure had no effect on [Ca²⁺]_i levels in cultured HUVECs and showed no modulation of the H₂O₂-, HlyA-, or thrombin-mediated increase in [Ca²⁺]_i. In addition, ADM-treated endothelial cells displayed no changes of intracellular cGMP content, even in cells with PDE2 inhibition to block cGMP degradation.

In summary, the data presented indicate that specific binding of ADM to endothelial cells elevated cAMP levels, blocked H₂O₂- and thrombin-related MLC-phosphorylation, and prevented endothelial cell contraction. ADM markedly reduced thrombin-, HlyA-, and H₂O₂-related endothelial hyperpermeability. Simultaneous inhibition of cAMP-degrading PDE3/4 and ADM treatment acted synergistically. Moreover, treatment of rabbit lungs with ADM reduced H₂O₂-induced edema formation and increased cAMP levels in lung perfusate. Thus, ADM has the potential of being a new therapeutic tool in systemic inflammatory reactions by stabilizing the endothelial barrier and preventing vascular leakage.

	Acknowledgments

 
This work was supported by the Deutsche Forschungsgemeinschaft (SFB 547/C2 to Dr Rascher and SFB 547/B6 to Dr Schütte) and the German Federal Research Ministry (BMBF) to Dr Suttorp (CAPNETZ/C4). Parts of this work will be included in the MD thesis of M. Krisp.

Received January 11, 2002; accepted August 28, 2002.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Fein AM, Calalang-Colucci MG. Acute lung injury and acute respiratory distress syndrome in sepsis and septic shock. Crit Care Clin. 2000; 16: 289–317.[CrossRef][Medline] [Order article via Infotrieve]
2. Stevens T, Garcia JG, Shasby DM, Bhattacharya J, Malik AB. Mechanisms regulating endothelial cell barrier function. Am J Physiol. 2000; 279: L419–L422.
3. Borbiev T, Verin AD, Shi S, Liu F, Garcia JG. Regulation of endothelial cell barrier function by calcium/calmodulin-dependent protein kinase II. Am J Physiol. 2001; 280: L983–L990.
4. Carbajal JM, Gratrix ML, Yu CH, Schaeffer RC Jr. ROCK mediates thrombin’s endothelial barrier dysfunction. Am J Physiol. 2000; 279: C195–C204.
5. Nieuw Amerongen GP, van Delft S, Vermeer MA, Collard JG, van Hinsbergh VW. Activation of RhoA by thrombin in endothelial hyperpermeability: role of Rho kinase and protein tyrosine kinases. Circ Res. 2000; 87: 335–340.[Abstract/Free Full Text]
6. Tinsley JH, De Lanerolle P, Wilson E, Ma W, Yuan SY. Myosin light chain kinase transference induces myosin light chain activation and endothelial hyperpermeability. Am J Physiol. 2000; 279: C1285–C1289.
7. Siflinger-Birnboim A, Goligorsky MS, Del Vecchio PJ, Malik AB. Activation of protein kinase C pathway contributes to hydrogen peroxide-induced increase in endothelial permeability. Lab Invest. 1992; 67: 24–30.[Medline] [Order article via Infotrieve]
8. Suttorp N, Weber U, Welsch T, Schudt C. Role of phosphodiesterases in the regulation of endothelial permeability in vitro. J Clin Invest. 1993; 91: 1421–1428.[Medline] [Order article via Infotrieve]
9. Suttorp N, Hippenstiel S, Fuhrmann M, Krüll M, Podzuweit T. Role of nitric oxide and phosphodiesterase isoenzyme II for reduction of endothelial hyperpermeability. Am J Physiol. 1996; 270: C778–C785.[Medline] [Order article via Infotrieve]
10. Suttorp N, Ehreiser P, Hippenstiel S, Fuhrmann M, Krüll M, Tenor H, Schudt C. Hyperpermeability of pulmonary endothelial monolayer: protective role of phosphodiesterase isoenzymes 3 and 4. Lung. 1996; 174: 181–194.[Medline] [Order article via Infotrieve]
11. Garcia JG, Verin AD, Schaphorst KL. Regulation of thrombin-mediated endothelial cell contraction and permeability. Semin Thromb Hemost. 1996; 22: 309–315.[Medline] [Order article via Infotrieve]
12. Suttorp N, Floer B, Schnittler H, Seeger W, Bhakdi S. Effects of Escherichia coli hemolysin on endothelial cell function. Infect Immun. 1990; 58: 3796–3801.[Abstract/Free Full Text]
13. Suttorp N, Fuhrmann M, Tannert-Otto S, Grimminger F, Bhadki S. Pore-forming bacterial toxins potently induce release of nitric oxide in porcine endothelial cells. J Exp Med. 1993; 178: 337–341.[Abstract/Free Full Text]
14. Lum H, Jaffe HA, Schulz IT, Masood A, RayChaudhury A, Green RD. Expression of PKA inhibitor (PKI) gene abolishes cAMP-mediated protection to endothelial barrier dysfunction. Am J Physiol. 1999; 277: C580–C588.[Medline] [Order article via Infotrieve]
15. Hirata Y, Mitaka C, Sato K, Nagura T, Tsunoda Y, Amaha K, Marumo F. Increased circulating adrenomedullin, a novel vasodilatory peptide, in sepsis. J Clin Endocrinol Metab. 1996; 81: 1449–1453.[Abstract]
16. Nikitenko LL, Smith DM, Hague S, Wilson CR, Bicknell R, Rees MC. Adrenomedullin and the microvasculature. Trends Pharmacol Sci. 2002; 23: 101–103.[CrossRef][Medline] [Order article via Infotrieve]
17. Ueda S, Nishio K, Minamino N, Kubo A, Akai Y, Kangawa K, Matsuo H, Fujimura Y, Yoshioka A, Masui K, Doi N, Murao Y, Miyamoto S. Increased plasma levels of adrenomedullin in patients with systemic inflammatory response syndrome. Am J Respir Crit Care Med. 1999; 160: 132–136.[Abstract/Free Full Text]
18. Wang P, Zhou M, Ba ZF, Cioffi WG, Chaudry IH. Up-regulation of a novel potent vasodilatory peptide adrenomedullin during polymicrobial sepsis. Shock. 1998; 10: 118–122.[Medline] [Order article via Infotrieve]
19. Wang P. Adrenomedullin in sepsis and septic shock. Shock. 1998; 10: 383–384.[Medline] [Order article via Infotrieve]
20. Hinson JP, Kapas S, Smith DM. Adrenomedullin, a multifunctional regulatory peptide. Endocr Rev. 2000; 21: 138–167.[Abstract/Free Full Text]
21. Samson WK. Adrenomedullin and the control of fluid and electrolyte homeostasis. Annu Rev Physiol. 1999; 61: 363–389.[CrossRef][Medline] [Order article via Infotrieve]
22. Ishimitsu T, Miyata A, Matsuoka H, Kangawa K. Transcriptional regulation of human adrenomedullin gene in vascular endothelial cells. Biochem Biophys Res Commun. 1998; 243: 463–470.[CrossRef][Medline] [Order article via Infotrieve]
23. Buhlmann N, Leuthauser K, Muff R, Fischer JA, Born W. A receptor activity modifying protein (RAMP)2-dependent adrenomedullin receptor is a calcitonin gene-related peptide receptor when coexpressed with human RAMP1. Endocrinology. 1999; 140: 2883–2890.[Abstract/Free Full Text]
24. Barker S, Kapas S, Corder R, Clark AJ. Adrenomedullin acts via stimulation of cyclic AMP and not via calcium signalling in vascular cells in culture. J Hum Hypertens. 1996; 10: 421–423.[Medline] [Order article via Infotrieve]
25. Shimekake Y, Nagata K, Ohta S, Kambayashi Y, Teraoka H, Kitamura K, Eto T, Kangawa K, Matsuo H. Adrenomedullin stimulates two signal transduction pathways, cAMP accumulation and Ca²⁺ mobilization, in bovine aortic endothelial cells. J Biol Chem. 1995; 270: 4412–4417.[Abstract/Free Full Text]
26. Shindo T, Kurihara H, Maemura K, Kurihara Y, Kuwaki T, Izumida T, Minamino N, Ju KH, Morita H, Oh-hashi Y, Kumada M, Kangawa K, Nagai R, Yazaki Y. Hypotension and resistance to lipopolysaccharide-induced shock in transgenic mice overexpressing adrenomedullin in their vasculature. Circulation. 2000; 101: 2309–2316.[Abstract/Free Full Text]
27. Caron KM, Smithies O. Extreme hydrops fetalis and cardiovascular abnormalities in mice lacking a functional adrenomedullin gene. Proc Natl Acad Sci U S A. 2001; 98: 615–619.[Abstract/Free Full Text]
28. Hippenstiel S, Soeth S, Kellas B, Fuhrmann O, Seybold J, Krüll M, Eichel-Streiber C, Goebeler M, Ludwig S, Suttorp N. Rho proteins and the p38-MAPK pathway are important mediators for LPS-induced interleukin-8 expression in human endothelial cells. Blood. 2000; 95: 3044–3051.[Abstract/Free Full Text]
29. Krüll M, Dold C, Hippenstiel S, Rosseau S, Lohmeyer J, Suttorp N. Escherichia coli hemolysin and Staphylococcus aureus -toxin potently induce neutrophil adhesion to cultured human endothelial cells. J Immunol. 1996; 157: 4133–4140.[Abstract]
30. Feth F, Rascher W, Michel MC. G-protein coupling and signalling of Y1-like neuropeptide Y receptors in SK-N-MC cells. Naunyn Schmiedebergs Arch Pharmacol. 1991; 344: 1–7.[CrossRef][Medline] [Order article via Infotrieve]
31. Hippenstiel S, Krüll M, Ikemann A, Risau W, Clauss M, Suttorp N. VEGF induces hyperpermeability by a direct action on endothelial cells. Am J Physiol. 1998; 274: L678–L684.[Medline] [Order article via Infotrieve]
32. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca²⁺ indicators with greatly improved fluorescence properties. J Biol Chem. 1985; 260: 3440–3450.[Abstract/Free Full Text]
33. Seeger W, Walmrath D, Grimminger F, Rosseau S, Schütte H, Kramer HJ, Ermert L, Kiss L. Adult respiratory distress syndrome: model systems using isolated perfused rabbit lungs. Methods Enzymol. 1994; 233: 549–584.[Medline] [Order article via Infotrieve]
34. Seeger W, Hansen T, Rössig R, Schmehl T, Schütte H, Kramer HJ, Walmrath D, Weissmann N, Grimminger F, Suttorp N. Hydrogen peroxide-induced increase in lung endothelial and epithelial permeability: effect of adenylate cyclase stimulation and phosphodiesterase inhibition. Microvasc Res. 1995; 50: 1–17.[CrossRef][Medline] [Order article via Infotrieve]
35. Nishio K, Akai Y, Murao Y, Doi N, Ueda S, Tabuse H, Miyamoto S, Dohi K, Minamino N, Shoji H, Kitamura K, Kangawa K, Matsuo H. Increased plasma concentrations of adrenomedullin correlate with relaxation of vascular tone in patients with septic shock. Crit Care Med. 1997; 25: 953–957.[CrossRef][Medline] [Order article via Infotrieve]
36. Vijay P. Adrenomedullin in the treatment of pulmonary hypertension. Heart. 2000; 84: 575–576.[Free Full Text]
37. Garcia JG, Verin AD, Schaphorst K, Siddiqui R, Patterson CE, Csortos C, Natarajan V. Regulation of endothelial cell myosin light chain kinase by Rho, cortactin, and p60(src). Am J Physiol. 1999; 276: L989–L998.[Medline] [Order article via Infotrieve]
38. Essler M, Staddon JM, Weber PC, Aepfelbacher M. Cyclic AMP blocks bacterial lipopolysaccharide-induced myosin light chain phosphorylation in endothelial cells through inhibition of Rho/Rho kinase signaling. J Immunol. 2000; 164: 6543–6549.[Abstract/Free Full Text]
39. Comerford KM, Lawrence DW, Synnestvedt K, Levi BP, Colgan SP. Role of vasodilator-stimulated phosphoprotein in PKA-induced changes in endothelial junctional permeability. FASEB J. 2002; 16: 583–585.[Free Full Text]
40. Patterson CE, Davis HW, Schaphorst KL, Garcia JG. Mechanisms of cholera toxin prevention of thrombin- and PMA-induced endothelial cell barrier dysfunction. Microvasc Res. 1994; 48: 212–235.[CrossRef][Medline] [Order article via Infotrieve]
41. Ono Y, Okano I, Kojima M, Okada K, Kangawa K. Decreased gene expression of adrenomedullin receptor in mouse lungs during sepsis. Biochem Biophys Res Commun. 2000; 271: 197–202.[CrossRef][Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				S.-Q. Xu, K. Mahadev, X. Wu, L. Fuchsel, S. Donnelly, R. G. Scalia, and B. J. Goldstein Adiponectin Protects Against Angiotensin II or Tumor Necrosis Factor {alpha}-Induced Endothelial Cell Monolayer Hyperpermeability: Role of cAMP/PKA Signaling Arterioscler. Thromb. Vasc. Biol., May 1, 2008; 28(5): 899 - 905. [Abstract] [Full Text] [PDF]

				R. A. Oeckler and R. D. Hubmayr Ventilator-associated lung injury: a search for better therapeutic targets Eur. Respir. J., December 1, 2007; 30(6): 1216 - 1226. [Abstract] [Full Text] [PDF]

				B. Yanagawa, M. Kataoka, S. Ohnishi, M. Kodama, K. Tanaka, Y. Miyahara, H. Ishibashi-Ueda, Y. Aizawa, K. Kangawa, and N. Nagaya Infusion of adrenomedullin improves acute myocarditis via attenuation of myocardial inflammation and edema Cardiovasc Res, October 1, 2007; 76(1): 110 - 118. [Abstract] [Full Text] [PDF]

				T. Itoh, H. Obata, S. Murakami, K. Hamada, K. Kangawa, H. Kimura, and N. Nagaya Adrenomedullin ameliorates lipopolysaccharide-induced acute lung injury in rats Am J Physiol Lung Cell Mol Physiol, August 1, 2007; 293(2): L446 - L452. [Abstract] [Full Text] [PDF]

				C. Gibbons, R. Dackor, W. Dunworth, K. Fritz-Six, and K. M. Caron Receptor Activity-Modifying Proteins: RAMPing up Adrenomedullin Signaling Mol. Endocrinol., April 1, 2007; 21(4): 783 - 796. [Abstract] [Full Text] [PDF]

				V. Ramachandran, T. Arumugam, R. F. Hwang, J. K. Greenson, D. M. Simeone, and C. D. Logsdon Adrenomedullin Is Expressed in Pancreatic Cancer and Stimulates Cell Proliferation and Invasion in an Autocrine Manner via the Adrenomedullin Receptor, ADMR Cancer Res., March 15, 2007; 67(6): 2666 - 2675. [Abstract] [Full Text] [PDF]

				T. Yurugi-Kobayashi, H. Itoh, T. Schroeder, A. Nakano, G. Narazaki, F. Kita, K. Yanagi, M. Hiraoka-Kanie, E. Inoue, T. Ara, et al. Adrenomedullin/Cyclic AMP Pathway Induces Notch Activation and Differentiation of Arterial Endothelial Cells From Vascular Progenitors Arterioscler. Thromb. Vasc. Biol., September 1, 2006; 26(9): 1977 - 1984. [Abstract] [Full Text] [PDF]

				N. Schwarz, D. Renshaw, S. Kapas, and J. P Hinson Adrenomedullin increases the expression of calcitonin-like receptor and receptor activity modifying protein 2 mRNA in human microvascular endothelial cells. J. Endocrinol., August 1, 2006; 190(2): 505 - 514. [Abstract] [Full Text] [PDF]

				E. Zudaire, S. Portal-Nunez, and F. Cuttitta The central role of adrenomedullin in host defense J. Leukoc. Biol., August 1, 2006; 80(2): 237 - 244. [Abstract] [Full Text] [PDF]

				P. D. N'Guessan, S. Hippenstiel, M. O. Etouem, J. Zahlten, W. Beermann, D. Lindner, B. Opitz, M. Witzenrath, S. Rosseau, N. Suttorp, et al. Streptococcus pneumoniae induced p38 MAPK- and NF-{kappa}B-dependent COX-2 expression in human lung epithelium Am J Physiol Lung Cell Mol Physiol, June 1, 2006; 290(6): L1131 - L1138. [Abstract] [Full Text] [PDF]

				H. Slevogt, B. Schmeck, C. Jonatat, J. Zahlten, W. Beermann, V. van Laak, B. Opitz, S. Dietel, P. D. N'Guessan, S. Hippenstiel, et al. Moraxella catarrhalis induces inflammatory response of bronchial epithelial cells via MAPK and NF-{kappa}B activation and histone deacetylase activity reduction Am J Physiol Lung Cell Mol Physiol, May 1, 2006; 290(5): L818 - L826. [Abstract] [Full Text] [PDF]

K. Miyashita, H. Itoh, H. Arai, T. Suganami, N. Sawada, Y. Fukunaga, M. Sone, K. Yamahara, T. Yurugi-Kobayashi, K. Park, et al. The Neuroprotective and Vasculo-Neuro-Regenerative Roles of Adrenomedullin in Ischemic Brain and Its Therapeutic Potential Endocrinology, Ap