Proliferation, Migration, and ERK Activation in Human Retinal Endothelial Cells through A_2B Adenosine Receptor Stimulation

Abstract

purpose. The nucleoside adenosine has been implicated in angiogenesis. A previous study demonstrated that activation of the A_2B adenosine receptor (AdoR) increases cAMP accumulation, cell proliferation, and VEGF expression in human retinal endothelial cells (HRECs). In the present study, the role of this receptor was further characterized by examination of the effects of the selective A_2B AdoR antagonists 3-N-propylxanthine (enprofylline) and 3-isobutyl-8-pyrrolidinoxanthine (IPDX) on AdoR-mediated HREC proliferation, capillary tube formation, and signal-transduction pathways.

methods. HRECs were exposed to the adenosine analogue 5′-N-ethylcarboxamido-adenosine (NECA) in the absence or presence of AdoR antagonists. Migration was measured using Boyden chambers. Proliferation was assessed by counting cells. Western analysis was used to assess extracellular signal-related kinase (ERK) and cAMP response element-binding protein (CREB) in cell lysates. The effect of AdoR activation on tube formation was studied using cells grown on a synthetic basement membrane matrix.

results. NECA induced proliferation in a concentration-dependent manner that was inhibited by enprofylline and IPDX. NECA stimulated chemotaxis in a concentration-dependent manner that was also blocked by both A_2B AdoR antagonists. NECA activated ERK and CREB in HRECs. Both A_2B AdoR antagonists diminished activation of ERK by NECA exposure. ERK activation was also blocked by the ERK-mitogen–activated protein kinase (MAPK) inhibitor PD98059, but not by the protein kinase A (PKA) inhibitor H-89. CREB activation was blocked by H-89, but not by PD98059, suggesting that ERK activation is independent of PKA. NECA enhanced tube formation on the matrix, whereas both A_2B AdoR antagonists attenuated this effect.

conclusions. The selective A_2B AdoR antagonists, enprofylline and IPDX, inhibited NECA-stimulated proliferation, ERK activation, cell migration, and capillary tube formation. A_2B AdoR inhibition may offer a way to inhibit retinal angiogenesis and provide a novel therapeutic approach to treatment of diseases associated with aberrant neovascularization, such as diabetic retinopathy and retinopathy of prematurity.

Purine nucleosides represent a primitive but universal signaling system that is capable of modulating numerous cellular functions, including proliferation, morphogenesis, and differentiation. ¹ The nucleoside adenosine is released in greater amounts as a result of hypoxia or other damage. ² Physiological responses to adenosine include coronary vasodilation, inhibition of platelet aggregation, bradycardia, atrioventricular (A-V) block and attenuation of the cardiostimulatory effects of catecholamines. ³ Adenosine can stimulate endothelial cells to alter their pattern of gene expression. ⁴ In the retina, adenosine is released during ischemia and has protective effects on neuronal cells. ⁵ High levels of adenosine are associated with areas of vasculogenesis in the normal neonatal dog retina and with sites of angiogenesis in the canine model of oxygen-induced retinopathy. ^{6 7}  

Adenosine can interact with at least four subtypes of G-protein–coupled receptors (AdoR), termed A₁, A_2A, A_2B, and A₃. ² These receptors are encoded by distinct genes and can be distinguished, based on their affinities for adenosine agonists and antagonists. In addition, these receptors are classified based on their mechanism of signal transduction. A₁ and A₃ AdoRs interact with pertussis toxin–sensitive G proteins of the G_{α i} and G_{α o} family to inhibit adenylate cyclase, stimulate phospholipases, activate protein kinase C, and increase calcium release from intracellular stores. The A_2A (high-affinity) and A_2B (low-affinity) AdoRs were initially characterized by their ability to stimulate adenylate cyclase through G_{α s} coupling. ⁸ Recent evidence suggests that the A_2B receptor can also stimulate phospholipase C activity through G_{α q}. ^{9 10}  

Endothelial cells are known to have active adenosine metabolism, characterized by a large capacity for uptake and release of the nucleoside. ¹¹ Moreover, adenosine can stimulate endothelial cells to alter the pattern of gene expression ⁴ and has been implicated in angiogenesis. ^{12 13} We have previously reported that the activation of A_2B AdoR increases VEGF mRNA and protein expression in human retinal endothelial cells (HRECs). ¹⁴ Adenosine also has a synergistic effect with VEGF on retinal endothelial cell migration and capillary morphogenesis in vitro. ¹⁵  

Downstream signal-transduction pathways activated by adenosine and supporting its mitogenic and prosurvival actions include members of the mitogen-activated protein kinase (MAPK) family of kinases: extracellular signal-regulated kinase (ERK), c-jun terminal kinase (JNK), and p38. ^{10 16 17} Activation of protein kinase A (PKA) in response to adenosine may also promote cell proliferation. ¹⁸ It is not clear, however, whether the cAMP-PKA pathway and the MAPK pathways are convergent or divergent in HRECs. 

Our previous report that adenosine’s effects on HRECs are mediated through the A_2B AdoR was based on indirect evidence by excluding the involvement of other AdoR subtypes (A₁, A_2A, and A₃). That is, we showed that the mitogenic effect of NECA is accompanied by an increase in cAMP, supporting that the receptor subtype involved is either the A_2A or the A_2B, not the A₁ or A₃. Furthermore, selective antagonism of A_2A receptors did not affect AdoR-mediated cell proliferation or VEGF production and hence established that the A_2B receptor mediated these responses. 

Enprofylline (3-N-propylxanthine) is the first known selective, although not particularly potent, A_2B antagonist. Enprofylline has been shown to be 22 times more selective for human A_2B than for human A₁, 5 times more than for human A_2A, and 6 times more than for human A₃ AdoRs. The recent development of a more selective A_2B antagonist prompted the present study to extend our initial observations. More than 100 mono-, di-, and trisubstituted xanthine compounds were examined to understand the structural requirements of a selective A_2B antagonist and led to the development of 3-isobutyl-8-pyrrolidinoxanthine (IPDX), ¹⁹ a novel selective A_2B antagonist. IPDX is a xanthine derivative with 8-pyrrolidino substituent without substituents in the 1 position. IPDX has improved A_2B selectivity compared with enprofylline. IPDX is 38 times more selective for human A_2B than for human A₁, 55 times more for human A_2A, and 82 times more for human A₃ AdoRs. Furthermore, it does not show significant 3′,5′-cyclic nucleotide phosphodiesterase inhibitory activity and thus may be considered a selective A_2B antagonist. ¹⁹  

In this study, we used enprofylline and IDPX to demonstrate directly the attenuation of the effect of NECA on different aspects of the angiogenic process mediated through the A_2B AdoR. We examined cell proliferation, cell migration, capillary tube formation, and signaling cascades that modulate endothelial cell proliferation. 

Materials and Methods

Reagents

Tissue culture medium and antibiotics were purchased from Invitrogen Life Technologies (Carlsbad, CA). Additional media supplements (insulin-transferrin-selenium, endothelial cell growth supplement) were from Sigma-Aldrich (St. Louis, MO). The synthetic basement membrane matrix (Matrigel) was from BD Biosciences (San Jose, CA). Adenosine deaminase (ADA), NECA, xanthine amine congener (XAC), 8-cyclopentyl-1,3-dipropylxanthine (CPX), enprofylline, and anti-pan-ERK antibody were purchased from Sigma-Aldrich. Anti-phospho-ERK and anti-phospho-cAMP response element-binding protein (CREB) antibodies and PD98059 were purchased from Cell Signaling Technology (Beverly, MA). Horseradish peroxidase (HRP)–conjugated secondary antibodies (donkey anti-rabbit or anti-mouse) and enhanced chemiluminescence (ECL) reagents were purchased from Amersham-Pharmacia Biotech (Piscataway, NJ). H-89 was purchased from Calbiochem (San Diego, CA). IPDX and ZM241385 were provided by two of the authors (IB and LB, respectively). 

Cell Culture

Primary cultures of HRECs were prepared and maintained as previously described, ²⁰ and cells in passages 3 to 6 were used in the studies. The identity of endothelial cells in cultures was validated by demonstrating endothelial cell incorporation of fluorescence-labeled acetylated LDL and by flow cytometry analysis, as previously described. ¹⁴ To maintain purity of HRECs, several precautionary steps were taken. HRECs were grown in plasma-derived serum, which is free of platelet-derived growth factor and does not promote the growth of pericytes (the contaminating cell type in these preparations). In addition, cultures of HRECs were exposed to trypsin for only 45 seconds before passage. Endothelial cells float away during this short trypsin treatment, whereas pericytes remain attached to the substrate. 

Proliferation Assay

HRECs were seeded at 10⁴ cells/cm² in 24-well plates and allowed to adhere overnight. Cells were washed in Hanks’ balanced salt solution, and the medium was replaced with serum- and growth supplement–free medium (SFM) containing 0.1% dimethyl sulfoxide (DMSO) to control for drug vehicle for 24 hours to induce cell-cycle arrest. Cells were washed again and pretreated with 1 U/ml ADA for 30 minutes. Cells were then exposed to NECA (10 μM) with or without enprofylline (10 μM) or IPDX (10 μM), which exhibit greater selectivity for the A_2B receptor than other available antagonists. ¹⁹ Controls were HRECs exposed to SFM containing 0.1% DMSO or normal growth medium. For the next 3 days at 24-hour intervals, replicate wells were treated with trypsin and the cells were collected and counted in a cell counter (Beckman Coulter, Inc., Fullerton, CA). Each condition was examined in triplicate in three separate experiments, with cells from different donors used for each experiment. 

Chemotaxis

Endothelial cell chemotaxis was measured in blind-well chemotaxis chambers (Neuro Probe, Inc., Gaithersburg, MD) as previously described. ²¹ Briefly, a single-cell suspension of endothelial cells (3.0 × 10³ cells/well) in SFM with 0.1% DMSO was prepared and treated with ADA (1 U/ml) for 30 minutes. Thirty microliters of this suspension was placed in each of 48 lower wells of the blind-well apparatus. The wells were overlaid with a porous (5-μm-diameter pore) polyvinyl- and pyrrolidone-free polycarbonate membrane (Nucleopore, Pleasanton, CA), coated with 0.1% dermal collagen. The cells were allowed to attach to the membrane by inverting the chamber for 2 hours. The chambers were then placed upright and exposed to NECA alone (10 μM), NECA combined with enprofylline (10 μM), IPDX (10 μM), or the nonselective AdoR antagonist XAC (10 μM) in a 50-μl volume. After incubation for 12 hours, the membrane was recovered and scraped free of cells on the attachment side. The remaining cells, those that had migrated through the pores, were fixed in methanol, stained with modified Wright’s stain, and counterstained with hematoxylin and eosin. The positive control was 10% fetal bovine serum and the negative control was 1% albumin. Chemokinesis, the nonoriented increase in cell migration in response to a stimulus, was measured by adding equal concentrations of NECA or NECA plus one of the antagonists to both the lower and upper chambers. Treatment conditions were examined in triplicate in three separate experiments, using cells derived from different donors. 

ERK and CREB Activation

To characterize the signaling pathway used by adenosine to mediate its proliferative effects, activation of ERK1, ERK2, and CREB was examined. In all cases, the cells were incubated overnight in SFM, then incubated for 30 minutes in SFM containing DMSO (0.1%) and ADA (1 U/ml). All subsequent treatments were in medium containing 1 U/ml ADA to minimize the effect of endogenous adenosine. 

To determine whether the intracellular signaling pathways that mediate the proliferation of HRECs by NECA depend on ERK or CREB phosphorylation, the cells were treated with the ERK-MAPK (MEK) inhibitor PD98059 (50 μM) or the PKA inhibitor H-89 (50 μM) for 30 minutes and then stimulated with increasing concentrations of NECA (1 nM to 10 μM). 

Unlike the other studies in this report, in which the time to end point was in the range of 12 to 72 hours, the expected response time for signaling molecule phosphorylation in response to AdoR stimulation was approximately 10 minutes or less. Thus, for the studies examining the effect of selective AdoR antagonism, the cells were preincubated for 10 minutes with antagonist before adding increasing concentrations of NECA (10 nM to 10 μM) to ensure binding of antagonist to AdoR. The antagonists used were 10 μM enprofylline, 10 μM IPDX, 20 nM CPX, and 50 nM ZM241385. 

Ten minutes after the addition of NECA, the cells were lysed in a buffer containing 10 μg/ml aprotinin, 20 μg/ml leupeptin, 1 μM E-64, 1 μM okadaic acid, 200 μM sodium pervanadate, 1 mM dithiothreitol (all from Sigma-Aldrich), 5 mM EDTA, and 25 mM Tris (pH 6.8). Protein concentrations were determined using the BCA method (Pierce Chemical Co., Rockford, IL). Equal concentrations of protein were diluted 1:1 in Laemmli buffer, and proteins were fractionated on 10% SDS-polyacrylamide gels. Parallel gels were stained with Coomassie blue to verify loading, sample integrity, and protein separation. Proteins were transferred from acrylamide gels to polyvinyl difluoride (PVDF) membranes for immunodetection. ²² Membranes blocked for 1 hour with 5% powdered milk in TTBS (25 mM Tris-HCL, 150 mM NaCl, and 0.05% Tween 20) were probed at room temperature with either anti-phospho-ERK or anti-phospho-CREB diluted to 1 μg/ml. Total ERK was detected with a polyclonal pan-ERK antibody diluted to 25 ng/ml. HRP-conjugated secondary antibody was used for detection at a concentration of 1 μg/ml. All antibody incubations were for 1 hour, and membranes were washed three times in TTBS between antibody incubations. Peroxidase activity was detected by using ECL and visualized on x-ray film (XAR-2; Eastman Kodak, Rochester, NY) using 30-second to 1-minute exposure times. All treatment conditions were evaluated in cells derived from two (for MEK and PKA inhibition) or three (for AdoR antagonism) different donors. 

Endothelial Cell Tube Formation Assay

Endothelial tube formation was assessed on synthetic basement membrane assay (Matrigel; BD Biosciences). Briefly, the matrix was thawed and kept at 4°C. Multiwell plates and pipette tips were chilled to −20°C, and the matrix gel (125 μl) was added to each well of a 48-well plate and allowed to harden for a minimum of 1 hour at 37°C. HRECs were dissociated enzymatically (2 minutes at 37°C in 0.25% trypsin-EDTA), centrifuged (300g, 5 minutes), and resuspended in SFM containing 1 U/ml ADA and 0.1% DMSO for 30 minutes. HRECs (3 × 10⁴ in 100 μl per well) were added to the plates, and then 100 μl SFM containing ADA and NECA with and without the A_2B inhibitor was added at two times final concentration, and the plates were incubated at 37°C. Wells were photographed 48 hours after plating. Identical fields in each well were photographed to minimize the possible variation due to variable cell density caused by the settling of cells. Photographs were digitized and image-analysis software (Image; Scion, Frederick, MD) was used to measure total tube length in a predefined, comparable area from each well. All conditions were tested in duplicate wells in three separate experiments using cells from different donors. 

Statistical Analysis

Data were analyzed by one-way analysis of variance (ANOVA) followed by Bonferroni t-test, using either untreated or NECA-treated cells as the determinant when appropriate and are reported as mean ± SE. P < 0.05 was deemed significant. 

Results

Cell Proliferation in Response to NECA and AdoR Antagonists

NECA (10 μM) induced a time-dependent increase in HREC proliferation as measured by cell counts, achieving approximately 80% of the density of cells exposed to normal growth medium for 3 days. Both of the selective A_2B AdoR antagonists tested, enprofylline at 10 μM and IPDX at 10 μM, completely blocked the proliferative effect of NECA when added concurrently with the analogue (Fig. 1A) . NECA-induced cell proliferation is significantly different from proliferation in unstimulated cells (P < 0.05, by ANOVA). 

Effect of NECA and AdoR Antagonists on HREC Migration

NECA stimulated HREC chemotaxis when measured in the Boyden chamber assay and increased migration in a concentration-dependent manner. The concurrent addition of NECA and the nonselective AdoR antagonist XAC (10 μM) abolished NECA-stimulated migration of HRECs (data not shown). Concurrent addition of NECA with either of the selective A_2B antagonists IPDX (10 μM) or enprofylline (10 μM) also antagonized the stimulatory effect of NECA on chemotaxis (Fig. 1B) . Neither NECA alone nor NECA in combination with the AdoR antagonists induced chemokinesis (data not shown). NECA-induced cell migration is significantly different from that of unstimulated cells (P < 0.05, by ANOVA). Similarly, inhibition of migration by either A_2B AdoR antagonist is significantly different from inhibition by NECA alone (P < 0.05, by ANOVA). 

Effect of NECA and AdoR Antagonists on Signaling Molecules

The MEK inhibitor PD98059 (50 μM) decreased HREC viability (Davis and Grant, unpublished observations, 2000) and abolished ERK activation (Fig. 2 , top). The PKA inhibitor H-89 (50 μM) increased basal ERK activation (Fig. 2 , top right) and did not inhibit NECA-stimulated ERK activation. 

When HRECs were pretreated with PD98059 or H-89 and assayed for active CREB by Western blot, PD98059 did not alter CREB activation. However, H-89 inhibited CREB phosphorylation (Fig. 2 , middle). The amount of protein loaded in each lane of the gel was constant, as is shown by the staining for total ERK (Fig. 2 , bottom). 

Pretreatment with enprofylline (10 μM) or IPDX (10 μM) reduced ERK activation resulting from NECA stimulation of HRECs, but neither antagonist completely abolished the response induced by NECA (Fig. 3A) . Pretreatment with the A₁-selective antagonist CPX (20 nM) reduced NECA-stimulated ERK activation, but pretreatment with the A_2A-selective antagonist ZM241385 (50 nM), alone or in combination with CPX, did not attenuate ERK activation by subsequent addition of NECA (Fig. 3B) . In addition, the presence of ZM241385 increased basal ERK activation and reversed the antagonistic properties of CPX. 

Effect of NECA on Endothelial Cell Tube Formation

Figure 4 shows representative photomicrographs of endothelial cell tube formation on basement membrane matrix (Matrigel; BD Biosciences) in the absence or presence of NECA alone or concurrently with AdoR antagonists. Some tube formation was evident after 48 hours with unstimulated control cells (Fig. 4A) . NECA (10 μM) treatment supported extensive tube formation (Fig. 4B) that was inhibited by 10μ M IPDX (Fig. 4C) . At 48 hours 10 μM enprofylline (Fig. 4D) inhibited tube formation, resulting in fewer tubes of shorter length. 

Total tube length was measured on digitized photographs as pixel length. NECA increased tube length more than twofold in comparison with in untreated cells (74.2 ± 2.4 vs 35.7 ± 1.6, respectively, P < 0.01 by ANOVA). The addition of either enprofylline or IPDX decreased but did not completely negate the NECA-induced tube length (53.7 ± 0.9 and 66 ± 1.2, respectively, both P < 0.01 by ANOVA). Neither A_2B AdoR antagonist alone had a significant effect on total tube length compared with untreated cells (data not shown). 

Discussion

Angiogenesis is a compensatory process in response to insufficient tissue oxygenation. ²³ In the retinas of patients with diabetes, homeostatic abnormalities lead to retinal nonperfusion and subsequent ischemia. Ischemia leads to the neovascularization and disruption of the normal retinal vasculature that is characteristic of proliferative diabetic retinopathy. ²⁴  

Adenosine is released by hypoxic tissue in large amounts. This nucleoside is an endothelial cell mitogen, linking the altered metabolism in oxygen-deprived cells to the formation of new capillaries. ^{25 26} Several investigators have reported the mitogenic and proliferative effects of adenosine on cultured endothelial cells. ^{11 17 18 27 28} Previous studies have shown that adenosine-induced proliferation is mimicked by AdoR agonists and blocked by antagonists, thus implying an AdoR-mediated site of action. ^{14 18 27} In HRECs, the A_2B AdoR subtype that we previously localized using a specific antibody, is the primary receptor subtype responsible for mediating the increase in cAMP and VEGF expression as a result of exposure to the adenosine analogue NECA. ¹⁴  

In this study, the adenosine analogue NECA stimulated key phases relevant in angiogenesis, including cell migration (as assessed by Boyden chamber assay) and capillary tube formation (as assessed by the basement membrane matrix assay). NECA also stimulated signaling cascades associated with cell survival and proliferation. The selective A_2B antagonists enprofylline (with relative selectivity for A_2B 22 times more than for A₁, 5 times more than for A_2A, and 6 times more than for A₃) and IPDX (with relative selectivity for A_2B 38 times more than for A₁, 55 times more than for A_2A, and 82 times more than for A₃) attenuated or abolished these effects. 

Previous work has established that adenosine stimulates proliferation of endothelial cells but has not provided conclusive information about the receptor subtype(s) involved in this effect. Our findings and those of others ²⁹ are in contrast to the conclusions reached by Van Daele et al., ²⁸ who reported that only adenosine triphosphate (ATP) analogues (P-2 receptor agonists) and adenosine itself stimulate DNA synthesis in bovine endothelial cells. The differing results may be attributable to the species as well as the vascular bed studied. 

In the present study, the adenosine analogue NECA activated ERK and CREB in HRECs through a receptor that is partially antagonized by enprofylline and IPDX, implicating the A_2B receptors in mediating these effects. The incomplete antagonism of the ERK response suggests either involvement of a second receptor population or incomplete antagonism of the A_2B receptor. CPX also partially blocked ERK activation by NECA, indicating that both A₁ and A_2B receptors are coupled to ERK in HRECs. The A₁ agonist CPA is also capable of stimulating ERK (data not shown), and the portion of the response that is not antagonized by enprofylline and IPDX may result from A₁ receptor activation. Our data indicate that the activation of ERK and CREB occurs through divergent pathways, in that inhibition of PKA blocked CREB activation but did not affect ERK activation, other than producing an increase in basal levels of active ERK. Similarly, antagonism of the A_2A receptor, which is coupled exclusively to cAMP production, increased basal ERK activation. These data suggest a role for the cAMP-PKA pathway in inhibition of ERK but do not support a role for the cAMP-PKA pathway in NECA-induced ERK activation. These findings are shown in schematic form in Figure 5 . 

Our data are in agreement with other reports showing cAMP-induced decreases in ERK activation ^{30 31 32} and are in contrast to results observed by Sexl et al., ²⁹ who demonstrated that activation of ERK by NECA is cAMP-dependent in umbilical vein endothelial cells. Together, these studies suggest endothelial cell populations are differentially responsive to NECA. Recent evidence suggests that the A_2B AdoR is coupled through both G_{α s} and G_{α q}, ^{5 6} and our data implicate a signaling pathway other than the cAMP-PKA pathway in NECA-stimulated ERK, but not CREB, activation. Further experiments are needed to fully dissect the signaling intermediates involved in NECA-stimulated ERK activation and inhibition. 

In this study, we have shown that the nonselective AdoR agonist NECA stimulated phases relevant to retinal angiogenesis such as cell proliferation, cell migration, and capillary tube formation, as well as signaling cascades associated with cell survival and proliferation. A_2B antagonists blocked these effects of NECA. Thus, our findings raise the possibility that selective adenosine A_2B AdoR antagonists may attenuate the endothelial cell proliferation that leads to the aberrant angiogenesis seen in diabetic retinopathy. Consequently, A_2B AdoR antagonists may represent novel therapeutic approaches to modulate aberrant retinal neovascular responses. 

 Supported by National Institutes of Health Grants EY012601 and EY007739.

 Submitted for publication December 13, 2000; revised March 30, 2001; accepted April 26, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Figure 1.

 

(A) Time course of 10 μM NECA-induced HREC proliferation. The proliferative effect of NECA was completely blocked by either of the A_2B-specific antagonists enprofylline (10μ M) or IPDX (10 μM). (B) NECA (0.1–30 μM) induced a concentration-dependent increase in HREC migration compared with unstimulated cells (in SFM). Both enprofylline (10 μM) and IPDX (10μ M) blocked this effect. NECA-induced proliferation and migration are significantly different from that in unstimulated cells (P < 0.05, by ANOVA).

Figure 2.

 

Analysis of the intracellular signal-transduction pathways stimulated by NECA. Cells were pretreated with ADA (1 U/ml) in the presence or absence of either H-89 (50 μM) or PD98059 (50 μM) for 30 minutes before stimulation with NECA (1 nM to 10 μM) for 10 minutes. ERK activation was analyzed by Western blot using an antibody specific for the phosphorylated form of the enzyme. Phosphorylation (activation) of the transcription factor CREB was also analyzed with a phosphospecific antibody that recognizes the active form of the transcription factor. Blots were stripped and reprobed with a pan-ERK antibody to verify equal loading. Similar results were obtained using cells isolated from three donors.

Figure 3.

 

(A) Both enprofylline and IPDX antagonized ERK activation through NECA stimulation of AdoR. Cells were treated with 1 U/ml ADA for 30 minutes and preincubated with the antagonists for 10 minutes before stimulation with increasing concentrations of NECA for 10 minutes. Control cells were treated with vehicle (0.1% DMSO in SFM). Active ERK was detected by Western blot using an antibody against the phosphorylated form of the enzyme. Similar results were obtained in cells cultured from three different donors. (B) ERK activation by NECA was partially blocked by the A₁ AdoR antagonist CPX, but not by the A_2A antagonist ZM241385. HRECs were incubated for 10 minutes in the presence of 0.1% DMSO (control), 20 nM CPX, 50 nM ZM241385, or both CPX and ZM241385 and stimulated for 10 minutes with 10 nM, 1 mM, or 10 mM NECA. ERK activation was measured by Western blot with an antibody specific for the phosphorylated form of the enzyme. Total ERK levels were measured by Western blot in a parallel experiment. Data are representative of two experiments performed on cultured cells derived from two donors.

Figure 4.

 

Photomicrographs show that endothelial cell tube formation on basement membrane matrix was affected by 10 μM NECA. All micrographs were taken at 48 hours after treatment and were typical of results in cells from three donors. Total tube length was measured on digitized photographs as pixel length. Unstimulated control cells (A) showed some tube formation by 48 hours. At 48 hours NECA supported extensive tube formation (B). NECA increased tube length more than twofold more than in untreated cells (74.2 ± 2.4 vs 35.7 ± 1.6; P < 0.01). By contrast, the A_2B-selective antagonists IPDX at 10 μM (C) diminished NECA-induced tube formation (74.2 ± 2.4 vs 66 ± 1.2; P < 0.01). Enprofylline at 10 μM (D) diminished NECA-induced tube formation (74.2 ± 2.4 vs 53.7 ± 0.9, P < 0.01).

Figure 5.

 

The cell-signaling pathways thought to be involved in AdoR-mediated endothelial cell proliferation. Phosphorylation of the two major kinases involved in cell proliferation, CREB and ERK, appears to depend on divergent pathways. When HRECs were pretreated with the PKA inhibitor H-89 and then exposed to NECA, there was a reduction in CREB phosphorylation compared with NECA alone, indicating some (albeit minor) involvement of the G_αs-coupled A_2A AdoR through increased adenylyl cyclase activity. It also appears that PKA may be involved in downregulating ERK phosphorylation in HRECs, possibly through some pathway involving intermediates upstream of MEK, because inhibition of PKA by H-89 increased basal ERK phosphorylation. Arguing against a major role for the A_2A AdoR in HREC proliferation is the absence of effect of the A_2A-selective antagonist ZM241385 on proliferation (shown in previous studies) and on ERK activation (the current study). This latter observation further supports the belief that the A_2B AdoR is coupled to G_aq, because ERK phosphorylation was inhibited by both of the A_2B-selective antagonists enprofylline and IPDX and by the MEK inhibitor PD98059, but not by ZM241385. The A₁ AdoR may also play a slight role in modulating ERK activity through some unknown pathway, because the A₁-selective antagonist CPX appears to have diminished slightly the amount of ERK phosphorylation.