| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EDITORIAL FOCUS
Department of Physiology and Membrane Biology, School of Medicine, University of California, Davis, California
LESS THAN TEN YEARS HAS PASSED since the protein angiopoietin-1 (Ang-1) was identified as a ligand for Tie-2, a tyrosine kinase receptor found principally on vascular endothelial cells (8). Ang-1 and its receptor, in concert with VEGF and its associated receptors, are essential for vasculogenesis during embryonic development. Transgenic mice in which either Ang-1 or Tie-2 is disrupted fail to develop fully functional cardiovascular systems, and either knockout induces embryonic lethality (17, 18). Understanding the coordinated actions of VEGF and Ang-1 during blood vessel development and during angiogenesis in the adult is hoped to lead to therapeutic regulation of angiogenesis in a variety of clinically relevant states including promotion of vessel growth in wound healing or inhibition of angiogenesis in solid tumors.
A second action of Ang-1 is modulation the permeability of the vasculature. A new study by Baffert et al. in this issue (2) is the most recent of a series to investigate the actions of Ang-1 on vascular permeability of the microcirculation. The results of these and other studies suggest that Ang-1 and Tie-2 may be fruitful targets for therapeutic intervention under conditions in which vascular permeability is increased.
Transgenic mouse studies indicate that Ang-1 promotes a low basal permeability and can also reduce the response of endothelial cells to inflammatory stimulation. In mice overexpressing either VEGF or Ang-1 in skin, blood vessels are more numerous. Whereas VEGF overexpression induces vessels leaky to both plasma proteins and red blood cells, the vessels of animals overexpressing Ang-1 show no red blood cell extravasation and no sign of edema (19, 21). Moreover, the ear skin of mice transgenic for Ang-1 had reduced extravasation of Evans blue-labeled albumin when tested with inflammatory stimulators including mustard oil, serotonin, and platelet-activating factor (PAF). Even the response to subdermal injection of saline control solution was lower than in nontransgenic control individuals. To test the effects of Ang-1 overexpression in the adult an engineered version of the angiopoietin-1 gene was administered to adult mice as an adenoviral construct that resulted in transfection principally of hepatocytes and within 24 h yielded a high circulating concentration of the protein Ang-1* (2, 3, 20). Three days after transfection these animals were found to have a reduced sensitivity of the skin microvasculature to mustard oil-induced plasma leakage.
The techniques described above looked primarily at the efficacy of Ang-1* expression on preventing the inflammatory response to VEGF in transgene experiments or in comparison to Evans blue-labeled albumin extravasation from dermal vessels induced by topical application of mustard oil. The study by Baffert and colleagues in this issue (2) examines the ability of Ang-1* expression to protect a model of airway inflammation stimulated by acute injection of bradykinin (BK), an inflammatory mediator that has been used in many well-characterized studies of acute permeability induction. The authors observed accumulation of fluorescent microspheres within the walls of tracheal postcapillary venules within 3 min after BK injection in control animals without Ang-1* expression. Mice that had been transfected with Ang-1* 3 days earlier showed 69% less accumulation of microspheres. The reduced accumulation of microspheres was correlated with both fewer and smaller paracellular inflammatory gaps in the postcapillary venules during the initial minutes of inflammatory stimulation. Thus the paper by Baffert et al. (2) identifies a structural correlate to the effectiveness of Ang-1 and suggests that Ang-1 interferes with signaling mechanisms or with the structural rearrangements that are necessary to open inflammatory gaps. Ang-1 has become recognized as one of the few endogenous agents known to reduce permeability and able to moderate hyperpermeability characterized by leakage of plasma proteins through inflammatory gaps.
It is generally accepted that the primary pathway across the endothelium for plasma protein and water flow in inflamed tissue is via paracellular gaps formed between endothelial cells (1, 4, 13). Although the precise signaling pathways are far from well understood, a few mechanisms are known to block formation of inflammatory gaps. Increased intraendothelial cAMP has been used to block gap formation in a wide variety of tissues in vivo and in many cultured endothelial cell models. Intracellular concentration of cAMP can be induced by incubation of the cells with membrane-permeant analogs of cAMP, direct stimulation of adenylyl cyclase (AC), inhibition of phosphodiesterases, and agonist stimulation of G protein-coupled receptors linked to AC activation. The most widely quoted model of the cAMP mechanism suggests that cAMP blocks actin-myosin contraction through stimulation of PKA, with the assumption that contraction is a necessary step to open the endothelial gap. However, the evidence for that model has been challenged, and a new model in which cAMP stimulates endothelial adhesion independent of PKA has been proposed (7). An emerging concept is that strengthened adhesion between adjacent endothelial cells is an essential feature of the mode of action of agents that protect the endothelial barrier after exposure to injury or inflammatory conditions (7, 14, 22). The mechanisms that strengthen adhesion involve increased polymerization of actin in the endothelial cortical actin band, linkage of actin binding proteins to the cortical network, and tighter adhesion between molecular complexes associated with the actin band and molecules forming the adherens junction and the tight junction. In our laboratory we have recently demonstrated (23) that acute inhibition of the small GTPase Rac-1 causes large increases in the permeability of both venular microvessels in intact tissue and cultured endothelial monolayers. The increase in permeability is associated with a reduction of F-actin content by 50% and similar fractional reduction in VE-cadherin mediated adhesion of coated beads to endothelium. Also, preferential activation of Rac-1 by the bacterial toxin cytotoxic necrotizing factor-1 (CNF-1) significantly attenuated PAF induced increases in permeability. Our observation that activation of Rac-1 results in the protection of the endothelial barrier against acute inflammatory mediators is consistent with several recent observations that sphingosine 1-phosphate (S1P), which activates Rac-1 by a receptor-mediated mechanism, protects the lung permeability barrier both in vivo and in vitro (14). Similarly, S1P moderates the effect of PAF-induced inflammation in postcapillary venules in rat mesentery (15). Is it possible that Ang-1 and Tie-2 interact with these or similar adhesion promoting pathways?
Work with cultured endothelial cells confirms the broad picture that Ang-1 promotes endothelial barrier properties. It blocks Ca2+ entry, blocks RhoA activation, and blocks elevated monolayer permeability induced by VEGF, PAF, histamine, BK, and thrombin (9, 11, 12, 16). The specific signaling pathways are under investigation, but direct links from Tie-2 to the endothelial barrier are not known. Enhanced association of 
-catenin with VE-cadherin is reported with Ang-1 treatment, and their dissociation via VEGF stimulation is blocked by Ang-1, in agreement with the hypothesis that adhesion regulation is an important mode of permeability modulation. There are also reports that Ang-1 stimulation activates p21-activated kinase (PAK), one of the downstream effectors of Rac-1 (10). The latter studies demonstrate an association of endothelial cell motility with endothelial stimulation by Ang-1 and focus also on the generation of reactive oxygen species (ROS) and ROS mediation of endothelial migration, which is associated with impaired barrier function (10). The latter observations could possibly be reconciled with the stabilizing effects of Ang-1 on endothelium if regulation of Rac-1-dependent pathways varies in chronic inflammation compared with acute inflammatory stimulation of quiescent endothelium. Studies of the barrier-promoting effects of Ang-1 will benefit by further addressing regulation of adhesion structures.
One drawback of the Ang-1* transfection technique used by Baffert and colleagues (2) is that very high circulating levels of Ang-1* are induced (10 µg/ml) relative to the normal circulating concentration of 8 ng/ml (6). In part, the use of expression vectors was necessitated by the poor solubility and loss of activity of native Ang-1 when isolated and purified. Thus an important advance in Ang-1 research was the production of a soluble, stable, and highly potent engineered variant of the protein, called COMP-Ang-1 (5). COMP-Ang-1 stimulated endothelial cell migration and tube formation in culture and induced angiogenesis in a mouse corneal assay. The COMP-Ang-1-induced vessels were not leaky to fluorescent dextran, although this result needs to be further investigated and documented. Introduction of this modified protein will enable further testing of the effects of Tie-2 stimulation in both chronic and acute models of inflammation in vivo. With its increased solubility and stability, it will be interesting to learn whether stronger stimulation of Tie-2 results in more effective inflammatory blockade.
Development of pharmacological agonists and antagonists of Tie-2 would facilitate future advances. Can Tie-2 stimulation help reverse chronic inflammation without interfering with normal angiogenesis? Does it only inhibit the formation of large inflammatory gaps, or does it affect baseline permeability through action on the glycocalyx, tight junctions, vesiculovacuolar organelles, or fenestrations? Although many questions remain, the possibilities for Ang-1 and Tie-2 are both intriguing and promising.
GRANTS
This work was supported by National Heart, Lung, and Blood Institute Grant HL-28607.
FOOTNOTES
Address for reprint requests and other correspondence: R. Adamson, Dept. of Physiology and Membrane Biology, School of Medicine, Univ. of California, Davis, 1 Shields Ave., Davis, CA 95616
REFERENCES
in thrombin-induced endothelial permeability changes: inhibition by angiopoietin-1. Blood 104: 1716–1724, 2004.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
