Endostatin, the Proteolytic Fragment of Collagen XVIII, Induces Vasorelaxation
http://www.100md.com
D. Wenzel, A. Schmidt, K. Reimann, J. He
参见附件。
the Institute of Physiology I (D.W., B.K.F.), University of Bonn
Department of Molecular and Cellular Sport Medicine (A.S., W.B.), German Sport University Cologne
Institutes of Vegetative Physiology (K.R., G.P.) and Neurophysiology (J.H.), University of Cologne, Germany.
Abstract
Collagen XVIII is an important component of the extracellular matrix and is expressed in basement membranes. Its degradation results in the generation of endostatin claimed to possess antiangiogenic activity. To date, only limited knowledge exists with regard to the cellular signaling of this molecule. We show in single-cell measurements using the Ca2+ indicator fura-2 acetoxy methylester (fura-2 AM) and the nitric oxide (NO) indicator 4,5-diaminofluorescein diacetate that application of endostatin (ES) (5 pmol/L, 100 ng/mL) induced Ca2+ spikes and an increase of NO production in human and murine endothelial cells. The NO response was independent of an increase in cytosolic Ca2+ and blocked by the endothelial NO synthase (eNOS) inhibitor NG-nitro-L-arginine methyl ester and by incubation with pertussis toxin known to inhibit Gi/o proteins. The physiological relevance of this novel signaling pathway of ES was assessed with isometric force measurements in large and small arteries of mouse. Physiological concentrations of ES were found to decrease vascular tone in an endothelium-dependent manner. This occurred via an Arg-Gly-Asp (RGD) peptide–independent pathway through activation of Gi/o proteins, phosphatidylinositol 3-kinase, Akt, and eNOS. We conclude that the proteolytic matrix fragment ES is a prominent vasorelaxing agent. Because ES is constantly released into the blood, it is a novel regulator of blood pressure and, therefore, represents an interesting pharmacological target.
Key Words: vascular tone extracellular matrix endostatin nitric oxide G proteins
Introduction
The identification of novel endogenous and exogenous regulators of vascular tone is of great interest. Advances in this field may lead to a better understanding of existing pathophysiological conditions and reveal novel therapeutic targets. Besides the canonical agonists such as noradrenaline, acetylcholine, or histamine, components of the extracellular matrix (ECM) have been proposed to potentially play an important role in vasoregulation.1
Collagen XVIII is an ECM protein that is diffusely expressed in basement membranes and best known by its degradation product endostatin (ES). ES, a 20-kDa C-terminal fragment, is generated by cleavage of the ECM involving matrix metalloproteases, cathepsins,2 and elastases.3 ES is a potent angiogenesis inhibitor preventing neovascularization in vitro as well as in vivo. This effect is mediated by the inhibition of proliferation and migration and the enhancement of apoptosis in endothelial cells.4,5 In vivo, ES was reported to prevent tumor growth and metastasis,6 suggesting a promising therapeutic potential as anticancer agent. The biological effects of ES are mainly attributed to its antagonism to vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) signaling.7 ES was found to bind to the VEGF receptor kinase–inserted domain-containing receptor (KDR), thereby blocking tyrosine phosphorylation and the activation of mitogen-activated protein kinase (MAPK), which leads to an inhibition of VEGF-induced migration and proliferation.8
Recently, various signaling pathways have been reported to be affected by ES, but a coherent understanding of the cellular implications of ES has remained elusive to date. As nitric oxide (NO) is a critical signaling molecule in endothelial cells, the effect of ES on NO signaling was also determined. It was found to promote dephosphorylation of the endothelial NO synthase (eNOS) at Ser1177, preventing VEGF-induced NO synthesis and inhibiting endothelial cell migration.9
In the present study, we aimed to decipher ES-induced signaling pathways and their physiological relevance in endothelial cells using NO imaging and isometric force measurements. We found that physiological concentrations of ES relax large and small arteries via a NO-dependent, but Arg-Gly-Asp (RGD) peptide–independent, pathway.
Materials and Methods
Cell Culture and Magnetic Cell Sorting
Human umbilical vein endothelial cells (HUVEC) (PromoCell, Heidelberg, Germany) were cultivated in endothelial growth medium (PromoCell) and passaged every 5 to 7 days. Cells were maximally used for 8 passages. Embryonic stem cell derived endothelial cells (ESEC) were obtained from magnetic cell sorted embryoid bodies (EBs), which were cultivated as reported earlier.10 The magnetic cell sorting procedure was performed as previously described.11 Stem cells were cultivated in DMEM (Gibco) supplemented with 15% or 20% of FCS.
Single-Cell Cytosolic NO and Cytosolic Ca2+ Measurements
For NO measurements, HUVEC and ESEC were loaded with the cell-permeant NO indicator 4,5-diaminofluorescein diacetate (DAF-FM DA) (10 μmol/L) for 10 minutes at room temperature. Then, cells were washed with extracellular solution and perfusion was stopped to avoid shear stress. Ca2+ measurements were performed after 40 minutes of loading with the Ca2+ indicator fura-2 acetoxy methylester (fura-2 AM) (1 μmol/L). All measurements were performed at room temperature in normal extracellular solution containing (mmol/L): NaCl 140, KCl 5.4, CaCl2 3.6, MgCl2 1, Hepes 10, glucose 10, pH 7.4, which was also used as control solution. ES or KDR-inhibitor V1 was applied directly onto single cells using a patch pipette with a wide opening connected to a small perfusor syringe; substances were released at 0.5 μL/s. Fluorescence measurements were performed as described earlier,12 as excitation wavelength 485 nm for NO or 340/380 nm for Ca2+ measurements were chosen. The emitted fluorescence from the cells was imaged through a 535/50 nm emission filter. For the different conditions, fluorescence intensities after 15 minutes of recording were normalized to the initial level NO (F/F0)=1, whereas 340/380 ratio was shown for the Ca2+ measurements. Data analysis was performed using the TILLVision 4.0 software (TILL Photonics, Munich, Germany) and Sigmaplot (Jandel Scientific, San Rafael, Calif).
Nitrite/Nitrate Assay
For Nitrite/Nitrate Assay, HUVEC were incubated with Tyrode’s solution or ES (200 ng/mL) dissolved in Tyrode’s solution for 30 minutes. The supernatants were collected and immediately used for Nitrite/Nitrate Assay (catalog no. 04508; Fluka, Steinheim, Germany). The kit was used as described by the manufacturer. Fluorescence measurements were performed with a Tecan fluorimeter (Tecan AG, Mnnedorf, Switzerland).
Immunohistochemistry
HUVEC or mouse aorta were incubated with control Tyrode’s solution or ES (100 ng/mL) dissolved in Tyrode’s solution for 2, 5, 10, or 20 minutes, respectively. After fixation in 4% paraformaldehyde, the aorta was sucrose embedded and stored at –80°C. For immunohistochemistry, HUVEC or 10-μm sections of aorta were treated with the primary antibody anti-eNOS (SA-201, pAb, 1:1000; Biomol, Hamburg, Germany), and diaminobenzidine (DAB) staining was performed as described previously.13
Measurements of Contractile Tension
All animal work was performed in compliance with procedures approved by the institutional animal care committee. Wild-type mice of the strain HIM OF 1 or eNOS knockout (KO) mice of the inbred strain C57/Bl6 (The Jackson Laboratory, Bar Harbor, Me) (5 to 8 weeks old) were killed by cervical dislocation. The thoracic aorta or the tail artery were dissected free of connective tissue and cut into 2-mm-long segments in low Ca2+ physiological salt solution (PSS) containing (mmol/L): NaCl 118, KCl 5, Na2HPO4 1.2, MgCl2 1.2, CaCl2 0.16, Hepes 24, glucose 10, pH 7.4, at room temperature. In some experiments, the vascular endothelium was removed by scratching with a mouse whisker. For isometric contraction experiments, the arteries were mounted on a wire myograph (Multi Myograph System-610 mol/L, Danish Myo Technology A/S, Aarhus, Denmark). Experiments were performed in 2 mL of PSS (1.6 mmol/L CaCl2), pH 7.4, saturated with 100% O2 at 37°C. This solution was also used as control solution in isometric force measurements. The vessels were stretched radially to their optimal lumen diameter corresponding to 90% of the passive diameter of the vessel at 100 mm Hg. Following equilibration for 25 minutes, maximal and submaximal (EC75) contraction of vascular rings was elicited with 10 μmol/L and 0.5 μmol/L norepinephrine (NE), respectively. The steady level of submaximal NE contraction (EC75) was taken as 1 (100%). Statistical analysis compared maximal relaxations after test substance stimulation. Vessels were considered denuded when acetylcholine (ACh) (10 μmol/L) caused a relaxation of <15%. The kinetics of vasorelaxation was investigated to determine t1/2, indicating the time at half-maximal relaxation. Experiments were performed with recombinant mouse ES14 (concentrations ranging from 100 to 200 ng/mL) and commercially available recombinant mouse ES (concentrations ranging from 200 to 400 ng/mL) evoking similar degrees of vasorelaxation. ES obtained from Sigma proved less stable, even in the –80°C freezer, because of a lower degree of purity and, therefore, with time, became less consistent in its biological action compared with the abovementioned sources.
Chemicals
The chemicals used were obtained from Sigma-Aldrich (Steinheim, Germany) (NG-nitro-L-arginine methyl ester [L-NAME], pertussis toxin [PTX], ACh, NE, ES), Calbiochem (San Diego, Calif) (RGD peptide, LY 294002, KDR-inhibitor V1, ES, BAPTA AM), Alexis (Grünberg, Germany) (1H-[1,2,4]oxidiazolo[4,3-a]quinoxaline-1-one [ODQ], DAF FM DA), and Molecular Probes (fura-2 AM).
Statistical Analysis
All data are presented as mean±SEM. Data analysis was performed using ANOVA with Bonferroni post hoc test and/or Student’s t test for paired or unpaired data. Significance was considered at a probability value of <0.05.
Results
ES Induces an Increase of Intracellular NO
To test whether ES signals via a NO-dependent pathway, single-cell measurements of cytosolic NO were performed with the fluorescence dye DAF FM DA (10 μmol/L) in HUVEC and ESEC. These cell lines were chosen as they correspond to different developmental stages and are well characterized. After application of 100 ng/mL ES, a slow but prominent rise (14% increase after 15 minutes) of the DAF fluorescence intensity compared with controls was noticed (1.06±0.02, n=29 versus 0.93±0.02, n=34 (control solution); P<0.001) (Figure 1A, 1B, and 1E). The ES-induced NO response was concentration dependent as 10 ng/mL was without effect (0.85±0.02, n=35, P<0.001 versus 100 ng/mL ES) (Figure 1A, inset). The specificity of the NO elevation induced by ES was proven by preincubation with L-NAME (100 μmol/L for 10 minutes), a nonspecific NOS inhibitor, which abolished the response to ES (0.93±0.02, n=19, P<0.002 versus 100 ng/mL ES without L-NAME) (Figure 1C and 1E). Under conditions without NO increase the DAF fluorescence intensity declined steadily. We also investigated the effect of ES on cytosolic Ca2+. In accordance to earlier work,15 ES evoked Ca2+ spikes in HUVEC. Interestingly, clamping of cytosolic Ca2+ with the fast Ca2+ buffer BAPTA AM (10 μmol/L) prevented detectable changes of cytosolic Ca2+ but did not inhibit the ES-induced rise in NO, suggesting a Ca2+-independent NO production (Figure 1B, insets) (n=10). To rule out that the observed elevation of cytosolic NO in response to ES was just caused by its binding to the VEGF receptor KDR, we applied the KDR-inhibitor V1 (250 μmol/L).16 Under these conditions, cytosolic NO remained unaltered (0.91±0.02, n=20, P<0.001 versus 100 ng/mL ES) (Figure 1E), proving that ES acted at a receptor different from the VEGF receptor KDR. We next investigated the downstream signaling of ES by preincubating HUVEC with the Gi/o protein blocker PTX (1 μg/mL for 12 hours). The ES-mediated increase of cytosolic NO was abrogated in PTX-treated endothelial cells (0.96±0.02, n=35, P<0.004 versus 100 ng/mL ES without PTX) (Figure 1D and 1E). Comparison of ES-induced NO increase after RGD preincubation and ES treatment alone revealed no difference (P=0.3), indicating that the effect of ES is RGD independent. To validate the biological relevance of our findings, we tested the effect of ES on NO in a different endothelial cell preparation, the ESEC. ES (100 ng/mL) also in this cell type evoked an increase of cytosolic NO (1.19±0.03, n=59). This was again inhibited by L-NAME (1.09±0.02, n=36, P<0.03 versus 100 ng/mL ES) and PTX (1.06±0.01, n=41, P<0.001 versus 100 ng/mL ES). Interestingly, ESEC proved more sensitive to ES than HUVEC, because 10 ng/mL ES already had a maximal effect (1.19±0.03, n=25 [10 ng/mL] versus 1.19±0.03, n=59 [100 ng/mL]).
The data obtained with single-cell fluorescence measurements were corroborated by the Griess assay in HUVEC. ES (200 ng/mL) enhanced the Nitrite/Nitrate production from 1425±91 U to 1971±245 U (n=5, P<0.05) (control: 1483±159 U to 1446±31 U; n=5).
To determine whether ES led to NO production by activating the eNOS isoform, HUVEC were immunostained with an antibody that detects the conformational change of activated eNOS17 (Figure 2A and 2B). A clear increase in the eNOS staining was detected 2, 5, and 20 minutes after exposure to ES (100 ng/mL, n=25) fully in line with our single-cell experiments. To determine in more detail the intracellular signaling mechanism whereby ES leads to NO production, we performed Western blotting experiments. These showed that 5 minutes of ES exposure leads to a clear-cut increase of Akt and eNOS phosphorylation at Ser1177 (Figure 2C), a site known to be phosphorylated via a Ca2+-independent mechanism.18 These results were further corroborated by immunohistochemistry and semiquantitative analysis using TV densitometry (data not shown). Thus, ES increases cytosolic NO via a KDR-independent, Gi/o protein–coupled pathway through activation of Akt and eNOS.
ES Decreases Vascular Tone
Our experiments on single cells revealed that NO plays a central role in ES signaling. Because NO is known to be a key regulator of vascular contractility, we examined the impact of ES on vascular tone in mouse aorta using isometric force measurements. Addition of control solution (50 μL) (Figure 2D) or the control peptide M2BP1 (200 ng/mL)11 did not change the tone in NE precontracted (0.5 μmol/L) murine aortic rings (0.99±0.01, n=5, and 0.95±0.02, n=10, respectively) (Figure 2F). By contrast, ES relaxed NE precontracted arteries in a concentration-dependent manner. The vasorelaxation amounted to 13% of the steady state for 100 ng/mL ES (0.87±0.05, n=5) and 25% for 200 ng/mL ES (0.75±0.04, n=7) (P<0.03 versus PSS or M2BP1) (Figure 2E and 2F), whereas in accordance with the single-cell experiments 10 ng/mL, ES did not suffice to relax the vessels (1.03±0.01; n=4). ACh (10 μmol/L) induced a similar decrease of the tone (29%) in aortic rings as 200 ng/mL ES (0.71±0.03, n=27 [ACh]; P<0.001 versus PSS or M2BP1) (Figure 2F). Although our single-cell experiments revealed that ES induces NO production in the vascular endothelium, we could not rule out a direct effect of ES on smooth muscle. Therefore, we determined next whether exposure of aortic rings to ES led to eNOS activation in the endothelium and/or the smooth muscle. As depicted in Figure 3A and 3B, the ES-treated aortas (100 ng/mL for 10 minutes) displayed a pronounced staining of the endothelium for eNOS as compared with controls (n=3) but none in the smooth muscle layer. This was confirmed in endothelium-denuded aortic rings, where a strongly reduced ACh- and ES-induced relaxation of only 11% (ACh: 0.89±0.02, n=5; P<0.02 versus aorta with endothelium) or only 4% (200 ng/mL ES: 0.96±0.02, n=5; P<0.004 versus aorta with endothelium) was noticed (Figure 3C and 3D). Thus, the ES-induced relaxation of aorta appears to be dependent on eNOS activation in the intact vascular endothelium.
To further confirm the critical involvement of NO, NE precontracted aortic rings were preincubated either with L-NAME (200 μmol/L) (Figure 4A) or ODQ (1 μmol/L), a soluble guanylyl-cyclase (sGC) inhibitor. L-NAME and ODQ further increased tension (1.92±0.27, n=6 [L-NAME]); 1.99±0.18, n=4 [ODQ]), likely because of the blockade of basal NO production in the vessel wall.19 Preincubation with either L-NAME or ODQ blocked the relaxing effect of ES (1.84±0.19, n=6, P=1.0 versus without ES [L-NAME]; 2.15±0.23, n=4, P=1.0 versus without ES [ODQ]) (Figure 4B). In another series of experiments, relaxation of the vessel was induced by ES and thereafter antagonized by L-NAME or ODQ. This reversed vasorelaxation and led to a powerful contraction (data not shown). To unequivocally illustrate that eNOS activation is responsible for the ES-induced vasorelaxation, we tested its effect in aortic rings of eNOS KO mice. In these, ES was without effect (0.97±0.01, n=4, P<0.01 versus ES in wild type mice) (Figure 4C). Hence, ES decreases vascular tone via the eNOS/NO/sGC signaling pathway. Although also ACh is well known to transmit vascular relaxation via NO,20 the time course of relaxation induced by ACh was substantially faster compared with ES, pointing to potential differences in the respective signaling pathways (ACh: t1/2=29±12 seconds, n=13; versus ES: t1/2=106±14 seconds, n=13; P<0.01).
The results of the single-cell experiments suggested Gi/o to be involved in ES signaling. Therefore, aortic rings were preincubated for 12 hours with PTX (1 μg/mL). Under these conditions, the ES-induced relaxation was strongly attenuated to 3% (0.97±0.02, n=4; P<0.004 versus without PTX) (Figure 4D). In contrast, the effect of ACh was not altered by preincubation with PTX: 27% (0.73±0.03, n=4, P>0.6 versus without PTX) (Figure 4E). Thus, PTX inhibits the response to ES but not to ACh indicating that different G proteins are involved.
Phosphatidylinositol 3-kinase (PI3K) has been reported to be linked to PTX-sensitive G proteins.21 Therefore, aortic rings were preincubated with the PI3K blocker LY 294002 (30 μmol/L), which is known to decrease vascular tone.22 The resulting vasorelaxation (0.51±0.11, n=4) remained unaltered by the subsequent addition of ES (0.47±0.10, n=4, P>0.8 versus without ES) (Figure 5A and 5B).
Integrins are known to bind to RGD-containing matrix products, thereby potentially modulating directly vascular tone. In our experiments, we found that RGD peptides (1 mmol/L) induced vasodilation (0.78±0.04, n=11) that was enhanced even by the subsequent addition of ES (0.52±0.10, n=11, P<0.03 versus without ES) (Figure 5C and 5D). Although the ES effect is assumed to be mediated via a surface receptor, we performed experiments using biotinylated ES and different exposure times. Streptavidin labeling of the biotinylated ES demonstrated that no ES was taken up into cytosol within 15 minutes (Figure I in the online data supplement available at http://circres.ahajournals.org). These data were confirmed by His-tag labeling of ES (data not shown), clearly proving that the observed ES effects were not mediated through intracellular accumulation and direct activation of intracellular signaling cascades.
We also evaluated the effect of repetitive application of ES on vascular tone (cumulative dose of 300 ng/mL). As depicted in Figure 5E, this procedure led to a dose-dependent and sustained increase of relaxation (n=2). Because regulation of arterial tone primarily occurs in small vessels, we also tested the effect of ES in precontracted rings obtained from murine tail arteries. Similar to our findings in the aorta, ES evoked a prominent dose-dependent relaxation of NE precontracted small arteries (n=3) (Figure 5F). As would be expected, ES was without effect in tail arteries of eNOS KO mice (data not shown).
Discussion
In the present study, we show that the ECM cleavage protein ES elevates cytosolic NO production in endothelial cells, evoking vascular relaxation. We observed these effects at concentrations of ES close to its endogenous levels (serum concentrations of ES amount to 100 to 300 ng/mL),14 suggesting that this proteolytic ECM fragment may be an important regulator of vascular tone even under physiological conditions. In fact, it is well known that ES is constantly present in the blood because of the steady turnover of the ECM.23
ES signaling occurred independent from the VEGF receptor KDR via activation of a PTX-sensitive G protein, PI3K, and Akt. Our single-cell, Western blotting, and isometric contraction experiments further proved that ES signaling resulted in NO generation. This is supported by our data and a recent report in which a rapid increase of eNOS phosphorylation at the critical Ser1179 (the human Ser1177) site was observed within the first 5 minutes after ES application and explains why NO production was found to occur in a Ca2+-independent manner; in addition, endostatin-mediated eNOS phosphorylation was also seen at the Ser116, Ser617, and Ser635. The concentrations of ES used in the other study were, however, far above physiological levels.24 Interestingly, different signaling pathways and biological outcomes are observed using either very low (10 ng/mL) or higher concentrations of ES. We did not find an increase of NO after exposing HUVEC to 10 ng/mL ES. By contrast, other less mature types of endothelial cells, such as the ESEC, appear more sensitive to ES. In these cells, low concentrations of ES are sufficient to obtain a prominent NO response, which could be related to their high-affinity binding to ES.25
Importantly, the eNOS-dependent NO generation that is elicited by ES is physiologically relevant, as it causes vasorelaxation of large and small arteries. Our functional data are supported by a strong local association of ES with elastic fibers of aorta in vivo.26 Repetitive application of ES induced a permanent dose-dependent vasorelaxing effect, indicating that the transient relaxation noticed after a single-dose application may be attributable to instability or degradation of ES itself or one or several of the signaling components involved. The transient kinetics of a single-dose ES effect is in line with the study of Li et al where the ES-evoked cGMP increase peaked at 10 minutes.24 This time dependency in the ES-mediated NO increase may explain the findings of Urbich et al, who reported a decrease of VEGF-induced eNOS phosphorylation after exposing HUVEC to ES for 1 hour.9 Deininger et al provide a more complex example of ES regulation, as NO is proposed to evoke ES release in endothelial cells and subsequent ES-induced lowering of the NO production.27 These different results can be explained by the disparate time course of the experiments, because we investigated the immediate response to ES, whereas Deininger et al used the adenoviral-based ES transduction of endothelial cells, which takes several days.
Our pharmacological data further suggest that the ES-induced vasorelaxation occurs through a PTX-sensitive G protein and a PI3K isoform. Earlier studies evidenced that the PI3K/Akt kinase pathway leading to eNOS phosphorylation and enhancement of NO production18 is an important cascade in endothelial cells. At present, it is unclear whether the ES-mediated G protein and PI3K signaling are interdependent. PI3K is known to be stimulated by G protein subunits28; nevertheless, a direct interaction of the G protein with the eNOS cannot be excluded. Therefore, additional work is required to determine the distinct PI3K isoform(s) involved in ES signaling as well as the identity and the regulatory role of the G protein and subunits.
Earlier work suggested a potential role of components of the ECM in the regulation of vascular tone. In these studies, matricryptic sites, biologically active fragments that are exposed after conformational or structural changes of the ECM, have been investigated. The data suggest that synthetic RGD peptides and proteolytic fragments of collagen type I (also containing RGD sites) are potentially involved in the control of vascular contractility.29 Although ES does not contain RGD motifs, ES was reported to compete with RGD cyclic peptide to bind to 51 integrin.30 However, in contrast to earlier studies, we found that ES transmitted its signal in an RGD-independent manner. In addition, Leu-Asp-Val (LDV) peptides, another matricryptic site, are also known to influence vascular contractility. Nevertheless, this pathway seems unlikely for ES, as these peptides promote vasoconstriction in an endothelium-independent manner.31.Our internalization studies and those of others32 corroborate the assumption that ES acts via a receptor-dependent, G protein–coupled pathway in endothelial cells. It is difficult to predict at this time whether the ES-mediated vasorelaxation also plays a role in angiogenesis, although local increase of blood flow can induce and support angiogenesis. This assumption would be in line with reports indicating that besides its well-described antiangiogenic effects, ES can also display proangiogenic activity.11,33
Altogether, we show here, for the first time, that ES, a specific degradation product of the ECM, is a powerful regulator of vascular tone. This finding may prove helpful to better explain the mechanisms underlying a variety of pathophysiological conditions. In fact, in early stages of aneurysm formation, elastases and matrix metalloproteinases were shown to inhibit Ca2+ entry, resulting in vascular dilation.34 Thus, ES, which is known to be released by these proteases, may aggravate aneurysm formation. In addition, the herewith reported vasodilative effect of ES fits nicely with the observation that Down syndrome patients carrying an additional copy of ES on chromosome 21 have increased ES blood levels35 and experience hypotonus. The vasorelaxing properties of ES could also be beneficial for antiangiogenic cancer therapy, as the combination of ES with Avastin, a VEGF-antibody, proved more effective in tumor therapy.36 ES may reduce high blood pressure, a typical side-effect observed during treatment with inhibitors of VEGF signaling. In basic pathological states, such as inflammation and wound healing, augmented local blood supply is required and an increased cleavage of ECM proteins occurs.37,38 Thus, future studies should be directed at determining the relevance of ES with respect to its effects on macro- and microcirculation under these situations.
Acknowledgments
This work was supported by grants of the Cologne Fortune Program (project 30/2002) (to D.W.) and the German Research Foundation (Deutsche Forschungsgemeinschaft) (Pf 226/10-1). We thank D. Metzler for excellent technical assistance.
Footnotes
Original received May 10, 2005; resubmission received December 27, 2005; revised resubmission received March 6, 2006; accepted March 21, 2006.
References
Martinez-Lemus LA, Wu X, Wilson E, Hill MA, Davis GE, Davis MJ, Meininger GA. Integrins as unique receptors for vascular control. J Vasc Res. 2003; 40: 211–233. [Order article via Infotrieve]
Felbor U, Dreier L, Bryant RA, Ploegh HL, Olsen BR, Mothes W. Secreted cathepsin L generates endostatin from collagen XVIII. EMBO J. 2000; 19: 1187–1194. [Order article via Infotrieve]
Wen W, Moses MA, Wiederschain D, Arbiser JL, Folkman J. The generation of endostatin is mediated by elastase. Cancer Res. 1999; 59: 6052–6056.
Dhanabal M, Ramchandran R, Waterman MJ, Lu H, Knebelmann B, Segal M, Sukhatme VP. Endostatin induces endothelial cell apoptosis. J Biol Chem. 1999; 274: 11721–11726.
Taddei L, Chiarugi P, Brogelli L, Cirri P, Magnelli L, Raugei G, Ziche M, Granger HJ, Chiarugi V, Ramponi G. Inhibitory effect of full-length human endostatin on in vitro angiogenesis. Biochem Biophys Res Commun. 1999; 263: 340–345. [Order article via Infotrieve]
O’Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, Flynn E, Birkhead JR, Olsen BR, Folkman J. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell. 1997; 88: 277–285. [Order article via Infotrieve]
Eriksson K, Magnusson P, Dixelius J, Claesson-Welsh L, Cross MJ. Angiostatin and endostatin inhibit endothelial cell migration in response to FGF and VEGF without interfering with specific intracellular signal transduction pathways. FEBS Lett. 2003; 536: 19–24. [Order article via Infotrieve]
Kim YM, Hwang S, Kim YM, Pyun BJ, Kim TY, Lee ST, Gho YS, Kwon YG. Endostatin blocks vascular endothelial growth factor-mediated signaling via direct interaction with KDR/Flk-1. J Biol Chem. 2002; 277: 27872–27879.
Urbich C, Reissner A, Chavakis E, Dernbach E, Haendeler J, Fleming I, Zeiher AM, Kaszkin M, Dimmeler S. Dephosphorylation of endothelial nitric oxide synthase contributes to the anti-angiogenic effects of endostatin. FASEB J. 2002; 16: 706–708.
Kazemi S, Wenzel D, Kolossov E, Lenka N, Raible A, Sasse P, Hescheler J, Addicks K, Fleischmann BK, Bloch W. Differential role of bFGF and VEGF for vasculogenesis. Cell Physiol Biochem. 2002; 12: 55–62. [Order article via Infotrieve]
Schmidt A, Wenzel D, Ferring I, Kazemi S, Sasaki T, Hescheler J, Timpl R, Addicks K, Fleischmann BK, Bloch W. Influence of endostatin on embryonic vasculo- and angiogenesis. Dev Dyn. 2004; 230: 468–480. [Order article via Infotrieve]
Freedman BD, Liu QH, Gaulton G, Kotlikoff MI, Hescheler J, Fleischmann BK. ATP-evoked Ca2+ transients and currents in murine thymocytes: possible role for P2X receptors in death by neglect. Eur J Immunol. 1999; 29: 1635–1646. [Order article via Infotrieve]
Pott C, Brixius K, Bundkirchen A, Bolck B, Bloch W, Steinritz D, Mehlhorn U, Schwinger RH. The preferential beta3-adrenoceptor agonist BRL 37344 increases force via beta1-/beta2-adrenoceptors and induces endothelial nitric oxide synthase via beta3-adrenoceptors in human atrial myocardium. Br J Pharmacol. 2003; 138: 521–529. [Order article via Infotrieve]
Sasaki T, Fukai N, Mann K, Gohring W, Olsen BR, Timpl R. Structure, function and tissue forms of the C-terminal globular domain of collagen XVIII containing the angiogenesis inhibitor endostatin. EMBO J. 1998; 17: 4249–4256. [Order article via Infotrieve]
Jiang L, Jha V, Dhanabal M, Sukhatme VP, Alper SL. Intracellular Ca(2+) signaling in endothelial cells by the angiogenesis inhibitors endostatin and angiostatin. Am J Physiol Cell Physiol. 2001; 280: C1140–C1150.
Binetruy-Tournaire R, Demangel C, Malavaud B, Vassy R, Rouyre S, Kraemer M, Plouet J, Derbin C, Perret G, Mazie JC. Identification of a peptide blocking vascular endothelial growth factor (VEGF)-mediated angiogenesis. EMBO J. 2000; 19: 1525–1533. [Order article via Infotrieve]
Bloch W, Mehlhorn U, Krahwinkel A, Reiner M, Dittrich M, Schmidt A, Addicks K. Ischemia increases detectable endothelial nitric oxide synthase in rat and human myocardium. Nitric Oxide. 2001; 5: 317–333. [Order article via Infotrieve]
Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher AM. Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature. 1999; 399: 601–605. [Order article via Infotrieve]
Boulanger CM, Heymes C, Benessiano J, Geske RS, Levy BI, Vanhoutte PM. Neuronal nitric oxide synthase is expressed in rat vascular smooth muscle cells: activation by angiotensin II in hypertension. Circ Res. 1998; 83: 1271–1278.
Ignarro LJ, Byrns RE, Buga GM, Wood KS. Endothelium-derived relaxing factor from pulmonary artery and vein possesses pharmacologic and chemical properties identical to those of nitric oxide radical. Circ Res. 1987; 61: 866–879.
Orr AW, Pallero MA, Murphy-Ullrich JE. Thrombospondin stimulates focal adhesion disassembly through Gi- and phosphoinositide 3-kinase-dependent ERK activation. J Biol Chem. 2002; 277: 20453–20460.
Su X, Smolock EM, Marcel KN, Moreland RS. Phosphatidylinositol 3-kinase modulates vascular smooth muscle contraction by calcium and myosin light chain phosphorylation-independent and -dependent pathways. Am J Physiol Heart Circ Physiol. 2004; 286: H657–H666.
Streuli C. Extracellular matrix remodelling and cellular differentiation. Curr Opin Cell Biol. 1999; 11: 634–640. [Order article via Infotrieve]
Li C, Harris MB, Venema VJ, Venema RC. Endostatin induces acute endothelial nitric oxide and prostacyclin release. Biochem Biophys Res Commun. 2005; 329: 873–878. [Order article via Infotrieve]
Schmidt A, Addicks K, Bloch W. Opposite effects of endostatin on different endothelial cells. Cancer Biol Ther. 2004; 3: 1162–1166. [Order article via Infotrieve]
Miosge N, Sasaki T, Timpl R. Angiogenesis inhibitor endostatin is a distinct component of elastic fibers in vessel walls. FASEB J. 1999; 13: 1743–1750.
Deininger MH, Wybranietz WA, Graepler FT, Lauer UM, Meyermann R, Schluesener HJ. Endothelial endostatin release is induced by general cell stress and modulated by the nitric oxide/cGMP pathway. FASEB. 2003; 17: 1267–1276.
Brock C, Schaefer M, Reusch HP, Czupalla C, Michalke M, Spicher K, Schultz G, Nurnberg B. Roles of G beta gamma in membrane recruitment and activation of p110 gamma/p101 phosphoinositide 3-kinase gamma. J Cell Biol. 2003; 160: 89–99.
Mogford JE, Davis GE, Platts SH, Meininger GA. Vascular smooth muscle alpha v beta 3 integrin mediates arteriolar vasodilation in response to RGD peptides. Circ Res. 1996; 79: 821–826.
Sudhakar A, Sugimoto H, Yang C, Lively J, Zeisberg M, Kalluri R. Human tumstatin and human endostatin exhibit distinct antiangiogenic activities mediated by alpha v beta 3 and alpha 5 beta 1 integrins. Proc Natl Acad Sci U S A. 2003; 100: 4766–4771.
Waitkus-Edwards KR, Martinez-Lemus LA, Wu X, Trzeciakowski JP, Davis MJ, Davis GE, Meininger GA. alpha(4)beta(1) Integrin activation of L-type calcium channels in vascular smooth muscle causes arteriole vasoconstriction. Circ Res. 2002; 90: 473–480.
Dixelius J, Larsson H, Sasaki T, Holmqvist K, Lu L, Engstrom A, Timpl R, Welsh M, Claesson-Welsh L. Endostatin-induced tyrosine kinase signaling through the Shb adaptor protein regulates endothelial cell apoptosis. Blood. 2000; 95: 3403–3411.
Morbidelli L, Donnini S, Chillemi F, Giachetti A, Ziche M. Angiosuppressive and angiostimulatory effects exerted by synthetic partial sequences of endostatin. Clin Cancer Res. 2003; 9: 5358–5369.
Chew DK, Conte MS, Khalil RA. Matrix metalloproteinase-specific inhibition of Ca2+ entry mechanisms of vascular contraction. J Vasc Surg. 2004; 40: 1001–1010. [Order article via Infotrieve]
Sund M, Hamano Y, Sugimoto H, Sudhakar A, Soubasakos M, Yerramalla U, Benjamin LE, Lawler J, Kieran M, Shah A, Kalluri R. Function of endogenous inhibitors of angiogenesis as endothelium-specific tumor suppressors. Proc Natl Acad Sci U S A. 2005; 102: 2934–2939.
Abdollahi A, Lipson KE, Sckell A, Zieher H, Klenke F, Poerschke D, Roth A, Han X, Krix M, Bischof M, Hahnfeldt P, Grone HJ, Debus J, Hlatky L, Huber PE. Combined therapy with direct and indirect angiogenesis inhibition results in enhanced antiangiogenic and antitumor effects. Cancer Res. 2003; 63: 8890–8898.
Krieglstein CF, Cerwinka WH, Sprague AG, Laroux FS, Grisham MB, Koteliansky VE, Senninger N, Granger DN, de Fougerolles AR. Collagen-binding integrin alpha1beta1 regulates intestinal inflammation in experimental colitis. J Clin Invest. 2002; 110: 1773–1782.
Li J, Zhang YP, Kirsner RS. Angiogenesis in wound repair: angiogenic growth factors and the extracellular matrix. Microsc Res Tech. 2003; 60: 107–114. [Order article via Infotrieve]
the Institute of Physiology I (D.W., B.K.F.), University of Bonn
Department of Molecular and Cellular Sport Medicine (A.S., W.B.), German Sport University Cologne
Institutes of Vegetative Physiology (K.R., G.P.) and Neurophysiology (J.H.), University of Cologne, Germany.
Abstract
Collagen XVIII is an important component of the extracellular matrix and is expressed in basement membranes. Its degradation results in the generation of endostatin claimed to possess antiangiogenic activity. To date, only limited knowledge exists with regard to the cellular signaling of this molecule. We show in single-cell measurements using the Ca2+ indicator fura-2 acetoxy methylester (fura-2 AM) and the nitric oxide (NO) indicator 4,5-diaminofluorescein diacetate that application of endostatin (ES) (5 pmol/L, 100 ng/mL) induced Ca2+ spikes and an increase of NO production in human and murine endothelial cells. The NO response was independent of an increase in cytosolic Ca2+ and blocked by the endothelial NO synthase (eNOS) inhibitor NG-nitro-L-arginine methyl ester and by incubation with pertussis toxin known to inhibit Gi/o proteins. The physiological relevance of this novel signaling pathway of ES was assessed with isometric force measurements in large and small arteries of mouse. Physiological concentrations of ES were found to decrease vascular tone in an endothelium-dependent manner. This occurred via an Arg-Gly-Asp (RGD) peptide–independent pathway through activation of Gi/o proteins, phosphatidylinositol 3-kinase, Akt, and eNOS. We conclude that the proteolytic matrix fragment ES is a prominent vasorelaxing agent. Because ES is constantly released into the blood, it is a novel regulator of blood pressure and, therefore, represents an interesting pharmacological target.
Key Words: vascular tone extracellular matrix endostatin nitric oxide G proteins
Introduction
The identification of novel endogenous and exogenous regulators of vascular tone is of great interest. Advances in this field may lead to a better understanding of existing pathophysiological conditions and reveal novel therapeutic targets. Besides the canonical agonists such as noradrenaline, acetylcholine, or histamine, components of the extracellular matrix (ECM) have been proposed to potentially play an important role in vasoregulation.1
Collagen XVIII is an ECM protein that is diffusely expressed in basement membranes and best known by its degradation product endostatin (ES). ES, a 20-kDa C-terminal fragment, is generated by cleavage of the ECM involving matrix metalloproteases, cathepsins,2 and elastases.3 ES is a potent angiogenesis inhibitor preventing neovascularization in vitro as well as in vivo. This effect is mediated by the inhibition of proliferation and migration and the enhancement of apoptosis in endothelial cells.4,5 In vivo, ES was reported to prevent tumor growth and metastasis,6 suggesting a promising therapeutic potential as anticancer agent. The biological effects of ES are mainly attributed to its antagonism to vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) signaling.7 ES was found to bind to the VEGF receptor kinase–inserted domain-containing receptor (KDR), thereby blocking tyrosine phosphorylation and the activation of mitogen-activated protein kinase (MAPK), which leads to an inhibition of VEGF-induced migration and proliferation.8
Recently, various signaling pathways have been reported to be affected by ES, but a coherent understanding of the cellular implications of ES has remained elusive to date. As nitric oxide (NO) is a critical signaling molecule in endothelial cells, the effect of ES on NO signaling was also determined. It was found to promote dephosphorylation of the endothelial NO synthase (eNOS) at Ser1177, preventing VEGF-induced NO synthesis and inhibiting endothelial cell migration.9
In the present study, we aimed to decipher ES-induced signaling pathways and their physiological relevance in endothelial cells using NO imaging and isometric force measurements. We found that physiological concentrations of ES relax large and small arteries via a NO-dependent, but Arg-Gly-Asp (RGD) peptide–independent, pathway.
Materials and Methods
Cell Culture and Magnetic Cell Sorting
Human umbilical vein endothelial cells (HUVEC) (PromoCell, Heidelberg, Germany) were cultivated in endothelial growth medium (PromoCell) and passaged every 5 to 7 days. Cells were maximally used for 8 passages. Embryonic stem cell derived endothelial cells (ESEC) were obtained from magnetic cell sorted embryoid bodies (EBs), which were cultivated as reported earlier.10 The magnetic cell sorting procedure was performed as previously described.11 Stem cells were cultivated in DMEM (Gibco) supplemented with 15% or 20% of FCS.
Single-Cell Cytosolic NO and Cytosolic Ca2+ Measurements
For NO measurements, HUVEC and ESEC were loaded with the cell-permeant NO indicator 4,5-diaminofluorescein diacetate (DAF-FM DA) (10 μmol/L) for 10 minutes at room temperature. Then, cells were washed with extracellular solution and perfusion was stopped to avoid shear stress. Ca2+ measurements were performed after 40 minutes of loading with the Ca2+ indicator fura-2 acetoxy methylester (fura-2 AM) (1 μmol/L). All measurements were performed at room temperature in normal extracellular solution containing (mmol/L): NaCl 140, KCl 5.4, CaCl2 3.6, MgCl2 1, Hepes 10, glucose 10, pH 7.4, which was also used as control solution. ES or KDR-inhibitor V1 was applied directly onto single cells using a patch pipette with a wide opening connected to a small perfusor syringe; substances were released at 0.5 μL/s. Fluorescence measurements were performed as described earlier,12 as excitation wavelength 485 nm for NO or 340/380 nm for Ca2+ measurements were chosen. The emitted fluorescence from the cells was imaged through a 535/50 nm emission filter. For the different conditions, fluorescence intensities after 15 minutes of recording were normalized to the initial level NO (F/F0)=1, whereas 340/380 ratio was shown for the Ca2+ measurements. Data analysis was performed using the TILLVision 4.0 software (TILL Photonics, Munich, Germany) and Sigmaplot (Jandel Scientific, San Rafael, Calif).
Nitrite/Nitrate Assay
For Nitrite/Nitrate Assay, HUVEC were incubated with Tyrode’s solution or ES (200 ng/mL) dissolved in Tyrode’s solution for 30 minutes. The supernatants were collected and immediately used for Nitrite/Nitrate Assay (catalog no. 04508; Fluka, Steinheim, Germany). The kit was used as described by the manufacturer. Fluorescence measurements were performed with a Tecan fluorimeter (Tecan AG, Mnnedorf, Switzerland).
Immunohistochemistry
HUVEC or mouse aorta were incubated with control Tyrode’s solution or ES (100 ng/mL) dissolved in Tyrode’s solution for 2, 5, 10, or 20 minutes, respectively. After fixation in 4% paraformaldehyde, the aorta was sucrose embedded and stored at –80°C. For immunohistochemistry, HUVEC or 10-μm sections of aorta were treated with the primary antibody anti-eNOS (SA-201, pAb, 1:1000; Biomol, Hamburg, Germany), and diaminobenzidine (DAB) staining was performed as described previously.13
Measurements of Contractile Tension
All animal work was performed in compliance with procedures approved by the institutional animal care committee. Wild-type mice of the strain HIM OF 1 or eNOS knockout (KO) mice of the inbred strain C57/Bl6 (The Jackson Laboratory, Bar Harbor, Me) (5 to 8 weeks old) were killed by cervical dislocation. The thoracic aorta or the tail artery were dissected free of connective tissue and cut into 2-mm-long segments in low Ca2+ physiological salt solution (PSS) containing (mmol/L): NaCl 118, KCl 5, Na2HPO4 1.2, MgCl2 1.2, CaCl2 0.16, Hepes 24, glucose 10, pH 7.4, at room temperature. In some experiments, the vascular endothelium was removed by scratching with a mouse whisker. For isometric contraction experiments, the arteries were mounted on a wire myograph (Multi Myograph System-610 mol/L, Danish Myo Technology A/S, Aarhus, Denmark). Experiments were performed in 2 mL of PSS (1.6 mmol/L CaCl2), pH 7.4, saturated with 100% O2 at 37°C. This solution was also used as control solution in isometric force measurements. The vessels were stretched radially to their optimal lumen diameter corresponding to 90% of the passive diameter of the vessel at 100 mm Hg. Following equilibration for 25 minutes, maximal and submaximal (EC75) contraction of vascular rings was elicited with 10 μmol/L and 0.5 μmol/L norepinephrine (NE), respectively. The steady level of submaximal NE contraction (EC75) was taken as 1 (100%). Statistical analysis compared maximal relaxations after test substance stimulation. Vessels were considered denuded when acetylcholine (ACh) (10 μmol/L) caused a relaxation of <15%. The kinetics of vasorelaxation was investigated to determine t1/2, indicating the time at half-maximal relaxation. Experiments were performed with recombinant mouse ES14 (concentrations ranging from 100 to 200 ng/mL) and commercially available recombinant mouse ES (concentrations ranging from 200 to 400 ng/mL) evoking similar degrees of vasorelaxation. ES obtained from Sigma proved less stable, even in the –80°C freezer, because of a lower degree of purity and, therefore, with time, became less consistent in its biological action compared with the abovementioned sources.
Chemicals
The chemicals used were obtained from Sigma-Aldrich (Steinheim, Germany) (NG-nitro-L-arginine methyl ester [L-NAME], pertussis toxin [PTX], ACh, NE, ES), Calbiochem (San Diego, Calif) (RGD peptide, LY 294002, KDR-inhibitor V1, ES, BAPTA AM), Alexis (Grünberg, Germany) (1H-[1,2,4]oxidiazolo[4,3-a]quinoxaline-1-one [ODQ], DAF FM DA), and Molecular Probes (fura-2 AM).
Statistical Analysis
All data are presented as mean±SEM. Data analysis was performed using ANOVA with Bonferroni post hoc test and/or Student’s t test for paired or unpaired data. Significance was considered at a probability value of <0.05.
Results
ES Induces an Increase of Intracellular NO
To test whether ES signals via a NO-dependent pathway, single-cell measurements of cytosolic NO were performed with the fluorescence dye DAF FM DA (10 μmol/L) in HUVEC and ESEC. These cell lines were chosen as they correspond to different developmental stages and are well characterized. After application of 100 ng/mL ES, a slow but prominent rise (14% increase after 15 minutes) of the DAF fluorescence intensity compared with controls was noticed (1.06±0.02, n=29 versus 0.93±0.02, n=34 (control solution); P<0.001) (Figure 1A, 1B, and 1E). The ES-induced NO response was concentration dependent as 10 ng/mL was without effect (0.85±0.02, n=35, P<0.001 versus 100 ng/mL ES) (Figure 1A, inset). The specificity of the NO elevation induced by ES was proven by preincubation with L-NAME (100 μmol/L for 10 minutes), a nonspecific NOS inhibitor, which abolished the response to ES (0.93±0.02, n=19, P<0.002 versus 100 ng/mL ES without L-NAME) (Figure 1C and 1E). Under conditions without NO increase the DAF fluorescence intensity declined steadily. We also investigated the effect of ES on cytosolic Ca2+. In accordance to earlier work,15 ES evoked Ca2+ spikes in HUVEC. Interestingly, clamping of cytosolic Ca2+ with the fast Ca2+ buffer BAPTA AM (10 μmol/L) prevented detectable changes of cytosolic Ca2+ but did not inhibit the ES-induced rise in NO, suggesting a Ca2+-independent NO production (Figure 1B, insets) (n=10). To rule out that the observed elevation of cytosolic NO in response to ES was just caused by its binding to the VEGF receptor KDR, we applied the KDR-inhibitor V1 (250 μmol/L).16 Under these conditions, cytosolic NO remained unaltered (0.91±0.02, n=20, P<0.001 versus 100 ng/mL ES) (Figure 1E), proving that ES acted at a receptor different from the VEGF receptor KDR. We next investigated the downstream signaling of ES by preincubating HUVEC with the Gi/o protein blocker PTX (1 μg/mL for 12 hours). The ES-mediated increase of cytosolic NO was abrogated in PTX-treated endothelial cells (0.96±0.02, n=35, P<0.004 versus 100 ng/mL ES without PTX) (Figure 1D and 1E). Comparison of ES-induced NO increase after RGD preincubation and ES treatment alone revealed no difference (P=0.3), indicating that the effect of ES is RGD independent. To validate the biological relevance of our findings, we tested the effect of ES on NO in a different endothelial cell preparation, the ESEC. ES (100 ng/mL) also in this cell type evoked an increase of cytosolic NO (1.19±0.03, n=59). This was again inhibited by L-NAME (1.09±0.02, n=36, P<0.03 versus 100 ng/mL ES) and PTX (1.06±0.01, n=41, P<0.001 versus 100 ng/mL ES). Interestingly, ESEC proved more sensitive to ES than HUVEC, because 10 ng/mL ES already had a maximal effect (1.19±0.03, n=25 [10 ng/mL] versus 1.19±0.03, n=59 [100 ng/mL]).
The data obtained with single-cell fluorescence measurements were corroborated by the Griess assay in HUVEC. ES (200 ng/mL) enhanced the Nitrite/Nitrate production from 1425±91 U to 1971±245 U (n=5, P<0.05) (control: 1483±159 U to 1446±31 U; n=5).
To determine whether ES led to NO production by activating the eNOS isoform, HUVEC were immunostained with an antibody that detects the conformational change of activated eNOS17 (Figure 2A and 2B). A clear increase in the eNOS staining was detected 2, 5, and 20 minutes after exposure to ES (100 ng/mL, n=25) fully in line with our single-cell experiments. To determine in more detail the intracellular signaling mechanism whereby ES leads to NO production, we performed Western blotting experiments. These showed that 5 minutes of ES exposure leads to a clear-cut increase of Akt and eNOS phosphorylation at Ser1177 (Figure 2C), a site known to be phosphorylated via a Ca2+-independent mechanism.18 These results were further corroborated by immunohistochemistry and semiquantitative analysis using TV densitometry (data not shown). Thus, ES increases cytosolic NO via a KDR-independent, Gi/o protein–coupled pathway through activation of Akt and eNOS.
ES Decreases Vascular Tone
Our experiments on single cells revealed that NO plays a central role in ES signaling. Because NO is known to be a key regulator of vascular contractility, we examined the impact of ES on vascular tone in mouse aorta using isometric force measurements. Addition of control solution (50 μL) (Figure 2D) or the control peptide M2BP1 (200 ng/mL)11 did not change the tone in NE precontracted (0.5 μmol/L) murine aortic rings (0.99±0.01, n=5, and 0.95±0.02, n=10, respectively) (Figure 2F). By contrast, ES relaxed NE precontracted arteries in a concentration-dependent manner. The vasorelaxation amounted to 13% of the steady state for 100 ng/mL ES (0.87±0.05, n=5) and 25% for 200 ng/mL ES (0.75±0.04, n=7) (P<0.03 versus PSS or M2BP1) (Figure 2E and 2F), whereas in accordance with the single-cell experiments 10 ng/mL, ES did not suffice to relax the vessels (1.03±0.01; n=4). ACh (10 μmol/L) induced a similar decrease of the tone (29%) in aortic rings as 200 ng/mL ES (0.71±0.03, n=27 [ACh]; P<0.001 versus PSS or M2BP1) (Figure 2F). Although our single-cell experiments revealed that ES induces NO production in the vascular endothelium, we could not rule out a direct effect of ES on smooth muscle. Therefore, we determined next whether exposure of aortic rings to ES led to eNOS activation in the endothelium and/or the smooth muscle. As depicted in Figure 3A and 3B, the ES-treated aortas (100 ng/mL for 10 minutes) displayed a pronounced staining of the endothelium for eNOS as compared with controls (n=3) but none in the smooth muscle layer. This was confirmed in endothelium-denuded aortic rings, where a strongly reduced ACh- and ES-induced relaxation of only 11% (ACh: 0.89±0.02, n=5; P<0.02 versus aorta with endothelium) or only 4% (200 ng/mL ES: 0.96±0.02, n=5; P<0.004 versus aorta with endothelium) was noticed (Figure 3C and 3D). Thus, the ES-induced relaxation of aorta appears to be dependent on eNOS activation in the intact vascular endothelium.
To further confirm the critical involvement of NO, NE precontracted aortic rings were preincubated either with L-NAME (200 μmol/L) (Figure 4A) or ODQ (1 μmol/L), a soluble guanylyl-cyclase (sGC) inhibitor. L-NAME and ODQ further increased tension (1.92±0.27, n=6 [L-NAME]); 1.99±0.18, n=4 [ODQ]), likely because of the blockade of basal NO production in the vessel wall.19 Preincubation with either L-NAME or ODQ blocked the relaxing effect of ES (1.84±0.19, n=6, P=1.0 versus without ES [L-NAME]; 2.15±0.23, n=4, P=1.0 versus without ES [ODQ]) (Figure 4B). In another series of experiments, relaxation of the vessel was induced by ES and thereafter antagonized by L-NAME or ODQ. This reversed vasorelaxation and led to a powerful contraction (data not shown). To unequivocally illustrate that eNOS activation is responsible for the ES-induced vasorelaxation, we tested its effect in aortic rings of eNOS KO mice. In these, ES was without effect (0.97±0.01, n=4, P<0.01 versus ES in wild type mice) (Figure 4C). Hence, ES decreases vascular tone via the eNOS/NO/sGC signaling pathway. Although also ACh is well known to transmit vascular relaxation via NO,20 the time course of relaxation induced by ACh was substantially faster compared with ES, pointing to potential differences in the respective signaling pathways (ACh: t1/2=29±12 seconds, n=13; versus ES: t1/2=106±14 seconds, n=13; P<0.01).
The results of the single-cell experiments suggested Gi/o to be involved in ES signaling. Therefore, aortic rings were preincubated for 12 hours with PTX (1 μg/mL). Under these conditions, the ES-induced relaxation was strongly attenuated to 3% (0.97±0.02, n=4; P<0.004 versus without PTX) (Figure 4D). In contrast, the effect of ACh was not altered by preincubation with PTX: 27% (0.73±0.03, n=4, P>0.6 versus without PTX) (Figure 4E). Thus, PTX inhibits the response to ES but not to ACh indicating that different G proteins are involved.
Phosphatidylinositol 3-kinase (PI3K) has been reported to be linked to PTX-sensitive G proteins.21 Therefore, aortic rings were preincubated with the PI3K blocker LY 294002 (30 μmol/L), which is known to decrease vascular tone.22 The resulting vasorelaxation (0.51±0.11, n=4) remained unaltered by the subsequent addition of ES (0.47±0.10, n=4, P>0.8 versus without ES) (Figure 5A and 5B).
Integrins are known to bind to RGD-containing matrix products, thereby potentially modulating directly vascular tone. In our experiments, we found that RGD peptides (1 mmol/L) induced vasodilation (0.78±0.04, n=11) that was enhanced even by the subsequent addition of ES (0.52±0.10, n=11, P<0.03 versus without ES) (Figure 5C and 5D). Although the ES effect is assumed to be mediated via a surface receptor, we performed experiments using biotinylated ES and different exposure times. Streptavidin labeling of the biotinylated ES demonstrated that no ES was taken up into cytosol within 15 minutes (Figure I in the online data supplement available at http://circres.ahajournals.org). These data were confirmed by His-tag labeling of ES (data not shown), clearly proving that the observed ES effects were not mediated through intracellular accumulation and direct activation of intracellular signaling cascades.
We also evaluated the effect of repetitive application of ES on vascular tone (cumulative dose of 300 ng/mL). As depicted in Figure 5E, this procedure led to a dose-dependent and sustained increase of relaxation (n=2). Because regulation of arterial tone primarily occurs in small vessels, we also tested the effect of ES in precontracted rings obtained from murine tail arteries. Similar to our findings in the aorta, ES evoked a prominent dose-dependent relaxation of NE precontracted small arteries (n=3) (Figure 5F). As would be expected, ES was without effect in tail arteries of eNOS KO mice (data not shown).
Discussion
In the present study, we show that the ECM cleavage protein ES elevates cytosolic NO production in endothelial cells, evoking vascular relaxation. We observed these effects at concentrations of ES close to its endogenous levels (serum concentrations of ES amount to 100 to 300 ng/mL),14 suggesting that this proteolytic ECM fragment may be an important regulator of vascular tone even under physiological conditions. In fact, it is well known that ES is constantly present in the blood because of the steady turnover of the ECM.23
ES signaling occurred independent from the VEGF receptor KDR via activation of a PTX-sensitive G protein, PI3K, and Akt. Our single-cell, Western blotting, and isometric contraction experiments further proved that ES signaling resulted in NO generation. This is supported by our data and a recent report in which a rapid increase of eNOS phosphorylation at the critical Ser1179 (the human Ser1177) site was observed within the first 5 minutes after ES application and explains why NO production was found to occur in a Ca2+-independent manner; in addition, endostatin-mediated eNOS phosphorylation was also seen at the Ser116, Ser617, and Ser635. The concentrations of ES used in the other study were, however, far above physiological levels.24 Interestingly, different signaling pathways and biological outcomes are observed using either very low (10 ng/mL) or higher concentrations of ES. We did not find an increase of NO after exposing HUVEC to 10 ng/mL ES. By contrast, other less mature types of endothelial cells, such as the ESEC, appear more sensitive to ES. In these cells, low concentrations of ES are sufficient to obtain a prominent NO response, which could be related to their high-affinity binding to ES.25
Importantly, the eNOS-dependent NO generation that is elicited by ES is physiologically relevant, as it causes vasorelaxation of large and small arteries. Our functional data are supported by a strong local association of ES with elastic fibers of aorta in vivo.26 Repetitive application of ES induced a permanent dose-dependent vasorelaxing effect, indicating that the transient relaxation noticed after a single-dose application may be attributable to instability or degradation of ES itself or one or several of the signaling components involved. The transient kinetics of a single-dose ES effect is in line with the study of Li et al where the ES-evoked cGMP increase peaked at 10 minutes.24 This time dependency in the ES-mediated NO increase may explain the findings of Urbich et al, who reported a decrease of VEGF-induced eNOS phosphorylation after exposing HUVEC to ES for 1 hour.9 Deininger et al provide a more complex example of ES regulation, as NO is proposed to evoke ES release in endothelial cells and subsequent ES-induced lowering of the NO production.27 These different results can be explained by the disparate time course of the experiments, because we investigated the immediate response to ES, whereas Deininger et al used the adenoviral-based ES transduction of endothelial cells, which takes several days.
Our pharmacological data further suggest that the ES-induced vasorelaxation occurs through a PTX-sensitive G protein and a PI3K isoform. Earlier studies evidenced that the PI3K/Akt kinase pathway leading to eNOS phosphorylation and enhancement of NO production18 is an important cascade in endothelial cells. At present, it is unclear whether the ES-mediated G protein and PI3K signaling are interdependent. PI3K is known to be stimulated by G protein subunits28; nevertheless, a direct interaction of the G protein with the eNOS cannot be excluded. Therefore, additional work is required to determine the distinct PI3K isoform(s) involved in ES signaling as well as the identity and the regulatory role of the G protein and subunits.
Earlier work suggested a potential role of components of the ECM in the regulation of vascular tone. In these studies, matricryptic sites, biologically active fragments that are exposed after conformational or structural changes of the ECM, have been investigated. The data suggest that synthetic RGD peptides and proteolytic fragments of collagen type I (also containing RGD sites) are potentially involved in the control of vascular contractility.29 Although ES does not contain RGD motifs, ES was reported to compete with RGD cyclic peptide to bind to 51 integrin.30 However, in contrast to earlier studies, we found that ES transmitted its signal in an RGD-independent manner. In addition, Leu-Asp-Val (LDV) peptides, another matricryptic site, are also known to influence vascular contractility. Nevertheless, this pathway seems unlikely for ES, as these peptides promote vasoconstriction in an endothelium-independent manner.31.Our internalization studies and those of others32 corroborate the assumption that ES acts via a receptor-dependent, G protein–coupled pathway in endothelial cells. It is difficult to predict at this time whether the ES-mediated vasorelaxation also plays a role in angiogenesis, although local increase of blood flow can induce and support angiogenesis. This assumption would be in line with reports indicating that besides its well-described antiangiogenic effects, ES can also display proangiogenic activity.11,33
Altogether, we show here, for the first time, that ES, a specific degradation product of the ECM, is a powerful regulator of vascular tone. This finding may prove helpful to better explain the mechanisms underlying a variety of pathophysiological conditions. In fact, in early stages of aneurysm formation, elastases and matrix metalloproteinases were shown to inhibit Ca2+ entry, resulting in vascular dilation.34 Thus, ES, which is known to be released by these proteases, may aggravate aneurysm formation. In addition, the herewith reported vasodilative effect of ES fits nicely with the observation that Down syndrome patients carrying an additional copy of ES on chromosome 21 have increased ES blood levels35 and experience hypotonus. The vasorelaxing properties of ES could also be beneficial for antiangiogenic cancer therapy, as the combination of ES with Avastin, a VEGF-antibody, proved more effective in tumor therapy.36 ES may reduce high blood pressure, a typical side-effect observed during treatment with inhibitors of VEGF signaling. In basic pathological states, such as inflammation and wound healing, augmented local blood supply is required and an increased cleavage of ECM proteins occurs.37,38 Thus, future studies should be directed at determining the relevance of ES with respect to its effects on macro- and microcirculation under these situations.
Acknowledgments
This work was supported by grants of the Cologne Fortune Program (project 30/2002) (to D.W.) and the German Research Foundation (Deutsche Forschungsgemeinschaft) (Pf 226/10-1). We thank D. Metzler for excellent technical assistance.
Footnotes
Original received May 10, 2005; resubmission received December 27, 2005; revised resubmission received March 6, 2006; accepted March 21, 2006.
References
Martinez-Lemus LA, Wu X, Wilson E, Hill MA, Davis GE, Davis MJ, Meininger GA. Integrins as unique receptors for vascular control. J Vasc Res. 2003; 40: 211–233. [Order article via Infotrieve]
Felbor U, Dreier L, Bryant RA, Ploegh HL, Olsen BR, Mothes W. Secreted cathepsin L generates endostatin from collagen XVIII. EMBO J. 2000; 19: 1187–1194. [Order article via Infotrieve]
Wen W, Moses MA, Wiederschain D, Arbiser JL, Folkman J. The generation of endostatin is mediated by elastase. Cancer Res. 1999; 59: 6052–6056.
Dhanabal M, Ramchandran R, Waterman MJ, Lu H, Knebelmann B, Segal M, Sukhatme VP. Endostatin induces endothelial cell apoptosis. J Biol Chem. 1999; 274: 11721–11726.
Taddei L, Chiarugi P, Brogelli L, Cirri P, Magnelli L, Raugei G, Ziche M, Granger HJ, Chiarugi V, Ramponi G. Inhibitory effect of full-length human endostatin on in vitro angiogenesis. Biochem Biophys Res Commun. 1999; 263: 340–345. [Order article via Infotrieve]
O’Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, Flynn E, Birkhead JR, Olsen BR, Folkman J. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell. 1997; 88: 277–285. [Order article via Infotrieve]
Eriksson K, Magnusson P, Dixelius J, Claesson-Welsh L, Cross MJ. Angiostatin and endostatin inhibit endothelial cell migration in response to FGF and VEGF without interfering with specific intracellular signal transduction pathways. FEBS Lett. 2003; 536: 19–24. [Order article via Infotrieve]
Kim YM, Hwang S, Kim YM, Pyun BJ, Kim TY, Lee ST, Gho YS, Kwon YG. Endostatin blocks vascular endothelial growth factor-mediated signaling via direct interaction with KDR/Flk-1. J Biol Chem. 2002; 277: 27872–27879.
Urbich C, Reissner A, Chavakis E, Dernbach E, Haendeler J, Fleming I, Zeiher AM, Kaszkin M, Dimmeler S. Dephosphorylation of endothelial nitric oxide synthase contributes to the anti-angiogenic effects of endostatin. FASEB J. 2002; 16: 706–708.
Kazemi S, Wenzel D, Kolossov E, Lenka N, Raible A, Sasse P, Hescheler J, Addicks K, Fleischmann BK, Bloch W. Differential role of bFGF and VEGF for vasculogenesis. Cell Physiol Biochem. 2002; 12: 55–62. [Order article via Infotrieve]
Schmidt A, Wenzel D, Ferring I, Kazemi S, Sasaki T, Hescheler J, Timpl R, Addicks K, Fleischmann BK, Bloch W. Influence of endostatin on embryonic vasculo- and angiogenesis. Dev Dyn. 2004; 230: 468–480. [Order article via Infotrieve]
Freedman BD, Liu QH, Gaulton G, Kotlikoff MI, Hescheler J, Fleischmann BK. ATP-evoked Ca2+ transients and currents in murine thymocytes: possible role for P2X receptors in death by neglect. Eur J Immunol. 1999; 29: 1635–1646. [Order article via Infotrieve]
Pott C, Brixius K, Bundkirchen A, Bolck B, Bloch W, Steinritz D, Mehlhorn U, Schwinger RH. The preferential beta3-adrenoceptor agonist BRL 37344 increases force via beta1-/beta2-adrenoceptors and induces endothelial nitric oxide synthase via beta3-adrenoceptors in human atrial myocardium. Br J Pharmacol. 2003; 138: 521–529. [Order article via Infotrieve]
Sasaki T, Fukai N, Mann K, Gohring W, Olsen BR, Timpl R. Structure, function and tissue forms of the C-terminal globular domain of collagen XVIII containing the angiogenesis inhibitor endostatin. EMBO J. 1998; 17: 4249–4256. [Order article via Infotrieve]
Jiang L, Jha V, Dhanabal M, Sukhatme VP, Alper SL. Intracellular Ca(2+) signaling in endothelial cells by the angiogenesis inhibitors endostatin and angiostatin. Am J Physiol Cell Physiol. 2001; 280: C1140–C1150.
Binetruy-Tournaire R, Demangel C, Malavaud B, Vassy R, Rouyre S, Kraemer M, Plouet J, Derbin C, Perret G, Mazie JC. Identification of a peptide blocking vascular endothelial growth factor (VEGF)-mediated angiogenesis. EMBO J. 2000; 19: 1525–1533. [Order article via Infotrieve]
Bloch W, Mehlhorn U, Krahwinkel A, Reiner M, Dittrich M, Schmidt A, Addicks K. Ischemia increases detectable endothelial nitric oxide synthase in rat and human myocardium. Nitric Oxide. 2001; 5: 317–333. [Order article via Infotrieve]
Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher AM. Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature. 1999; 399: 601–605. [Order article via Infotrieve]
Boulanger CM, Heymes C, Benessiano J, Geske RS, Levy BI, Vanhoutte PM. Neuronal nitric oxide synthase is expressed in rat vascular smooth muscle cells: activation by angiotensin II in hypertension. Circ Res. 1998; 83: 1271–1278.
Ignarro LJ, Byrns RE, Buga GM, Wood KS. Endothelium-derived relaxing factor from pulmonary artery and vein possesses pharmacologic and chemical properties identical to those of nitric oxide radical. Circ Res. 1987; 61: 866–879.
Orr AW, Pallero MA, Murphy-Ullrich JE. Thrombospondin stimulates focal adhesion disassembly through Gi- and phosphoinositide 3-kinase-dependent ERK activation. J Biol Chem. 2002; 277: 20453–20460.
Su X, Smolock EM, Marcel KN, Moreland RS. Phosphatidylinositol 3-kinase modulates vascular smooth muscle contraction by calcium and myosin light chain phosphorylation-independent and -dependent pathways. Am J Physiol Heart Circ Physiol. 2004; 286: H657–H666.
Streuli C. Extracellular matrix remodelling and cellular differentiation. Curr Opin Cell Biol. 1999; 11: 634–640. [Order article via Infotrieve]
Li C, Harris MB, Venema VJ, Venema RC. Endostatin induces acute endothelial nitric oxide and prostacyclin release. Biochem Biophys Res Commun. 2005; 329: 873–878. [Order article via Infotrieve]
Schmidt A, Addicks K, Bloch W. Opposite effects of endostatin on different endothelial cells. Cancer Biol Ther. 2004; 3: 1162–1166. [Order article via Infotrieve]
Miosge N, Sasaki T, Timpl R. Angiogenesis inhibitor endostatin is a distinct component of elastic fibers in vessel walls. FASEB J. 1999; 13: 1743–1750.
Deininger MH, Wybranietz WA, Graepler FT, Lauer UM, Meyermann R, Schluesener HJ. Endothelial endostatin release is induced by general cell stress and modulated by the nitric oxide/cGMP pathway. FASEB. 2003; 17: 1267–1276.
Brock C, Schaefer M, Reusch HP, Czupalla C, Michalke M, Spicher K, Schultz G, Nurnberg B. Roles of G beta gamma in membrane recruitment and activation of p110 gamma/p101 phosphoinositide 3-kinase gamma. J Cell Biol. 2003; 160: 89–99.
Mogford JE, Davis GE, Platts SH, Meininger GA. Vascular smooth muscle alpha v beta 3 integrin mediates arteriolar vasodilation in response to RGD peptides. Circ Res. 1996; 79: 821–826.
Sudhakar A, Sugimoto H, Yang C, Lively J, Zeisberg M, Kalluri R. Human tumstatin and human endostatin exhibit distinct antiangiogenic activities mediated by alpha v beta 3 and alpha 5 beta 1 integrins. Proc Natl Acad Sci U S A. 2003; 100: 4766–4771.
Waitkus-Edwards KR, Martinez-Lemus LA, Wu X, Trzeciakowski JP, Davis MJ, Davis GE, Meininger GA. alpha(4)beta(1) Integrin activation of L-type calcium channels in vascular smooth muscle causes arteriole vasoconstriction. Circ Res. 2002; 90: 473–480.
Dixelius J, Larsson H, Sasaki T, Holmqvist K, Lu L, Engstrom A, Timpl R, Welsh M, Claesson-Welsh L. Endostatin-induced tyrosine kinase signaling through the Shb adaptor protein regulates endothelial cell apoptosis. Blood. 2000; 95: 3403–3411.
Morbidelli L, Donnini S, Chillemi F, Giachetti A, Ziche M. Angiosuppressive and angiostimulatory effects exerted by synthetic partial sequences of endostatin. Clin Cancer Res. 2003; 9: 5358–5369.
Chew DK, Conte MS, Khalil RA. Matrix metalloproteinase-specific inhibition of Ca2+ entry mechanisms of vascular contraction. J Vasc Surg. 2004; 40: 1001–1010. [Order article via Infotrieve]
Sund M, Hamano Y, Sugimoto H, Sudhakar A, Soubasakos M, Yerramalla U, Benjamin LE, Lawler J, Kieran M, Shah A, Kalluri R. Function of endogenous inhibitors of angiogenesis as endothelium-specific tumor suppressors. Proc Natl Acad Sci U S A. 2005; 102: 2934–2939.
Abdollahi A, Lipson KE, Sckell A, Zieher H, Klenke F, Poerschke D, Roth A, Han X, Krix M, Bischof M, Hahnfeldt P, Grone HJ, Debus J, Hlatky L, Huber PE. Combined therapy with direct and indirect angiogenesis inhibition results in enhanced antiangiogenic and antitumor effects. Cancer Res. 2003; 63: 8890–8898.
Krieglstein CF, Cerwinka WH, Sprague AG, Laroux FS, Grisham MB, Koteliansky VE, Senninger N, Granger DN, de Fougerolles AR. Collagen-binding integrin alpha1beta1 regulates intestinal inflammation in experimental colitis. J Clin Invest. 2002; 110: 1773–1782.
Li J, Zhang YP, Kirsner RS. Angiogenesis in wound repair: angiogenic growth factors and the extracellular matrix. Microsc Res Tech. 2003; 60: 107–114. [Order article via Infotrieve]
您现在查看是摘要介绍页，详见ORG附件。