Reduction in Preretinal Neovascularization by Ribozymes That Cleave the A2B Adenosine Receptor mRNA
Adenosine modulates a variety of cellular functions by interacting with specific cell surface G protein–coupled receptors (A1, A2A, A2B, and A3) and is a potential mediator of angiogenesis through the A2B receptor. The lack of a potent, selective A2B receptor inhibitor has hampered its characterization. Our goal was to design a hammerhead ribozyme that would specifically cleave the A2B receptor mRNA and examine its effect on retinal angiogenesis. Ribozymes specific for the mouse and human A2B receptor mRNAs were designed and cloned in expression plasmids. Human embryonic kidney (HEK) 293 cells were transfected with these plasmids and A2B receptor mRNA levels were determined by quantitative real-time RT-PCR. Human retinal endothelial cells (HRECs) were also transfected and cell migration was examined. The effects of these ribozymes on the levels of preretinal neovascularization were determined using a neonatal mouse model of oxygen-induced retinopathy (OIR). We produced a ribozyme with a Vmax of 515±125 pmol/min and a Kcat of 36.1±8.3 min−1 (P≤1×10−5). Transfection of HEK293 cells with the plasmid expressing the ribozyme reduced A2B receptor mRNA levels by 45±4.8% (P=5.1×10−5). Transfection of HRECs reduced NECA-stimulated migration of cells by 47.3±1.2% (P=7×10−4). Intraocular injection of the constructs into the mouse model reduced preretinal neovascularization by 53.5±8.2% (P=4.5×10−5). Our results suggest that the A2B receptor ribozyme will provide a tool for the selective inhibition of this receptor and provide further support for the role of A2B receptor in retinal angiogenesis.
Angiogenesis, the formation of new blood vessels from either preexisting vasculature or circulating stem cells, occurs as part of both normal development and pathologies of the retina. Retinal ischemia stimulates abnormal angiogenesis in conditions such as proliferative diabetic retinopathy (PDR) and retinopathy of prematurity (ROP). Substantial evidence supports a role for adenosine in promoting each step associated with angiogenesis.1,2 Adenosine can act as a mitogen in endothelial cells derived from numerous vascular beds to increase cell number,3–6 DNA synthesis,7 cell migration, and vascularization.8 Endothelial cells are known to have a very active adenosine metabolism.9 Adenosine is a critical mediator of blood flow changes in response to ischemia and is a significant component of the retina’s compensatory hyperemic response to ischemia, hypoxia, and hypoglycemia.10 In the retinal microvasculature, adenosine and adenosine analogues cause concentration-dependent vasodilation.11,12 These observations strongly support a role for endogenously released adenosine as a key mediator of retinal blood flow during conditions of reduced oxygen supply.11
Adenosine interacts with at least four subtypes of G protein–coupled receptors (A1, A2A, A2B, and A3),13,14 which are encoded by distinct genes and are differentiated based on their affinities for adenosine agonists and antagonists.15 The A1 and A2A (high affinity) receptors are activated by submicromolar concentrations of adenosine and the A2B and the A3 receptors (low affinity) are activated when adenosine levels rise to micromolar range.16 The receptors are also differentiated based on their signal transduction pathways.14 A1 and A3 receptors interact with the Gi and G0 family to inhibit adenylate cyclase, whereas A2A and A2B receptors interact with Gs and Gq to stimulate adenylate cyclase.15 Linden et al16a have shown that overexpression of A2B receptor results in cAMP accumulation and phospholipase C activation. On the other hand, A2B receptor antagonists (eg, enprofylline) inhibit agonist stimulated cAMP accumulation.17 In most cell types and organ systems, adenosine increases oxygen supply by activating the A2 receptors and decreases oxygen demand by activating the A1 receptor, demonstrating that adenosine may be a protective metabolite that rectifies imbalances between oxygen supply and demand.18
A2 receptors are associated with vessels and the A2A and A2B receptors have been implicated in angiogenesis. We have demonstrated that 5′-(N-ethylcarboxamido)adenosine (NECA) interacts with the A2B receptor in human retinal endothelial cells (HRECs) and stimulates cell migration and proliferation.19 NECA is a stable analogue of adenosine and activates all four adenosine receptors although with different potencies. Our findings and those of Sexl et al20 are in contrast with the work of Van Daele et al,21 who reported that adenosine stimulates only DNA synthesis in bovine aortic endothelial cells. These results suggest that endothelial cell populations differ in response to adenosine or adenosine analogues, and these differences may be attributed to differences in both the species and vascular beds studied. Takagi et al22 reported that endogenously released adenosine stimulates VEGF expression in bovine retinal endothelial cells and pericytes through stimulation of the A2A receptor. Lutty et al23 demonstrated that A2A receptors localized to the edge of the developing vasculature in canine retina. Taomoto et al24 also demonstrated high levels of A2A receptor immunoreactivity in immature intravitreal neovascular formations in the canine oxygen-induced retinopathy (OIR) model. Our in vitro studies, and others, using human cells, including HRECs,6 implicate the A2B receptor in release of VEGF, IL-8, and FGF-2. In addition, the results of our in vivo studies in a mouse model of OIR also demonstrate the role of the A2B receptor in retinal neovascularization.25
Hammerhead ribozymes are catalytic RNA molecules that cleave phosphodiester bonds within RNA nucleotides,26 and it has been proposed that ribozymes may act as potential inhibitors of cell proliferation.27 Several groups have used antisense RNA to control the expression of the A2B receptor in animals and in tissue culture.28–30 These results indicate several sites in the A2B receptor mRNA that are accessible to antisense oligoribonucleotides and, hence, to ribozyme cleavage. Because of their catalytic ability, a lower concentration of ribozyme molecules is required to achieve efficient inhibition of mRNA expression. Therefore, ribozymes should be more effective than antisense RNA in reducing the expression of the A2B receptor.31
Our goal was to target the A2B receptor with a hammerhead ribozyme to further demonstrate its role in angiogenesis. We designed and tested hammerhead ribozymes that specifically cleave the A2B adenosine receptor mRNA of mouse and human (Figure 1). We performed in vitro testing of the reaction kinetics of these ribozymes and determined the effect of these ribozymes on the migration of HRECs, and on the mRNA levels in HRECs and in human embryonic kidney (HEK) 293 cells. We characterized their effects on the level of preretinal neovascularization in the mouse model of OIR. Our results demonstrate that A2B receptor ribozymes are effective at reducing the levels of the A2B receptor mRNA and in inhibiting human retinal endothelial cell migration in vitro, as well as in reducing preretinal neovascularization in neonatal mice.
Materials and Methods
Synthetic RNA Targets and Ribozymes
We designed the A2B receptor ribozymes based on the mouse sequence of the A2B receptor (PubMed accession No. NM_007413). The sequences of theses ribozymes and targets are shown in Figure 1. RNA oligonucleotides for the active and inactive mouse A2B receptor hammerhead ribozymes and mouse and human targets were purchased from Dharmacon (Boulder, Colo) and deprotected following the manufacturer’s protocol. RNA oligonucleotides were 5′-end labeled with [γ-32P]-dATP (ICN) using polynucleotide kinase (Promega).
Time Course Analysis of Ribozyme Cleavage and Multiple-Turnover Kinetic Analysis
Time course analysis of ribozyme cleavage and multiple-turnover kinetic analysis were performed using the RNA oligonucleotides as previously described.32–34
Cloning of the Hammerhead Ribozymes Into the rAAV Expression Vector
Two complementary DNA oligonucleotides (Life Technologies) were annealed in order to produce a double-stranded DNA fragment coding for each hammerhead ribozyme. All DNA oligonucleotides were synthesized with 5′-phosphate groups. The DNA oligonucleotides were designed to generate a cut HindIII site at the 5′-end and a cut SpeI site at the 3′-end after annealing. The DNA oligonucleotides were incubated at 65°C for 2 minutes and annealed by slow cooling to room temperature for 30 minutes. The resulting double-stranded DNA fragment was ligated into the HindIII and SpeI sites of the rAAV vector pTRUF-21 (UF Vector Core, http://www.gtc.ufl.edu/gtc-home.htm). A self-cleaving hairpin ribozyme has been cloned downstream of the inserted hammerhead ribozymes into the SpeI and NsiI sites. This vector has the CMV/β-actin chimeric enhancer-promoter and results in the hairpin ribozyme cleaving eight bases downstream of the 3′-end of the hammerhead ribozymes. We term the vector with the hairpin ribozyme alone p21NewHp. We cloned genes for inactive ribozymes (Figure 1). and verified in time course experiments that these ribozymes were inactive (data not shown). The ligated plasmids were transformed into SURE electroporation competent cells (Stratagene) in order to maintain the integrity of the inverted terminal repeats. The ribozyme clones were verified by sequencing.
Transfection of HRECs
HRECs were transfected with plasmids expressing the A2B receptor ribozyme constructs as previously described.35,36
Relative Quantitative RT-PCR
Relative quantitative RT-PCR was performed on RNA isolated from HRECs transfected with plasmids expressing ribozymes (A2B Rz2 active and inactive) and p21NewHp. RNA was isolated from transfected HRECs using either the GenElute Direct mRNA Miniprep Kit (Sigma) for mRNA or the TRIzol Reagent (Invitrogen) for total RNA. Reverse transcription was accomplished using Superscript reverse transcriptase and a random hexamer (Invitrogen) according to manufacturer’s protocols.
PCR reactions to determine A2B receptor mRNA levels used gene-specific DNA oligonucleotides synthesized by Invitrogen (5′-GTACGTGGCGCTGGAGCTGG-3′ and 5′-CTTGCTCGGGT-CCCCGTGAC-3′). The linear range of the amplification of the A2B receptor RT-PCR product was determined by using a master PCR mix (1 μL RT product/ 50 μL, 200 μmol/L dNTPs, 1 mmol/L MgCl2, 0.4 μmol/L A2B oligonucleotides, 1X REDTaq DNA polymerase buffer [Sigma], 2 U REDTaq DNA polymerase [Sigma], and 0.5 μci/50 μL [α32P]-dATP [ICN]). This master mix was separated into eight 0.2 mL tubes, and amplification was performed with an annealing temperature of 61°C. Samples were removed at even-numbered cycles starting at cycle 26. For each PCR sample, 5 μL was removed and 2 μL of formamide dye mix was added. The reaction products were separated on a 6% polyacrylamide-8 mol/L urea gel. Dried gels were analyzed on a Molecular Dynamics PhosphoImager (Amersham) to determine the linear range of amplification. For this oligonucleotide pair, cycle number 34 was determined to be within the linear range of amplification and was used in subsequent experiments.
In the relative quantitative RT-PCR assays the level of A2B receptor mRNA was determined within each sample relative to an internal β-actin standard. β-actin mRNA levels were determined using the QuantumRNA β-actin primer/competimer oligonucleotide set from Ambion. The competimer oligonucleotide pair from the β-actin primer set anneals to the same targets as the primer oligonucleotide pair but they are blocked at their 3′-ends to prevent extension. This primer/competimer oligonucleotide set allowed us to determine the ratio of primer to competimer that yields a β-actin PCR fragment that is approximately equal molar to the A2B receptor PCR product. To determine the ratio of the primer/competimer oligonucleotide set required to achieve this, PCR reactions were performed as described above and amplification proceeded for 34 cycles. The ratio of primer to competimer oligonucleotide was determined to be 10:1 at a final concentration of 0.4 μmol/L for the combined primer/competimer mixture.
Migration Assays on Transfected HRECs
Transfected cells were assayed for their ability to migrate in response to NECA using a modified Boyden chamber assay as previously described.37 Cells (30 000 cells/well) were added to the bottom wells of modified Boyden chambers, which are then fully assembled with a collagen-coated, pyrrolidone- and pyrogen-free porous membrane. The cells adhere to the membrane for 4 hours at 37°C, after which 50 μL of the solution containing NECA was added to the wells. The chambers were then returned to 37°C for 12 hours to allow cell migration. The chambers were then disassembled and the membranes scraped on the side where the cells have adhered. The membranes are then fixed and stained with hematoxylin and eosin. Individuals masked to the identity of treatment counted the migrated cells. Control cells, transfected with p21NewHp, were compared with cells transfected with the ribozyme constructs. 10% FBS in DMEM was used as a positive control and DMEM alone as the negative control. Each test condition was assayed with a minimum of six replicate wells. Migrating cells were counted using a light microscope, and the number of migrating cells per well were calculated by averaging the number of nuclei counted in three separate, high-power (×400) fields. The values for the six replicate wells were then averaged. Each migration experiment was repeated a minimum of three times representing cultures derived from three donors.
Transfection of HEK293 Cells
HEK293 cells were grown to 90% confluence on 150-mm plates and transfected with the plasmid constructs using Lipofectamine 2000 reagent (Invitrogen) according to manufacturer’s protocol. The cells were harvested at 72 hours after transfection.
Messenger RNA was isolated from transfected HEK293 cells or HRECs using the GenElute Direct mRNA Miniprep Kit from Sigma. The cDNA was synthesized using either 2 or 4 μg of total RNA and TaqMan Reverse Transcription Reagents (PE Applied Biosystems) in 100 μL RT reaction. TaqMan real-time PCR analysis was applied using 1 μL cDNA per reaction and SYBR Green PCR Core Reagents on ABI Prism Sequence Detection System 5700 (PE Applied Biosystems). In each experiment, a standard curve for each primer pair was obtained using a serial dilution of total RNA samples prepared from cells that overexpressed A2B adenosine receptor. At the end of the PCR cycle, a dissociation curve was generated to ensure the amplification of a single product and the threshold cycle time (Ct values) for each gene was determined. Relative mRNA levels were calculated based on the Ct values and normalized to one of the following housekeeping genes: β-actin, cyclophilin, and ribosomal protein S9 (100%).
All statistical analysis was done using the Student t test in Microsoft Excel. Values of P<0.001 are indicated with an asterisk on the graphs.
All animal procedures used were in agreement with institutional guidelines and with the NIH Guide for the Care and Use of Laboratory Animals and approved by the University of Florida Institutional Animal Care and Use Committee. Six timed pregnant C57BL/6J mice were purchased from Jackson Laboratories (Bar Harbor, Maine). A total of 36 mouse pups were used for these studies.
Intraocular Injection Into the Mouse Model of Oxygen-Induced Retinopathy
In the neonatal mouse model of OIR, 7-day old mice were placed with their nursing dams in a 75% oxygen atmosphere for 5 days.38 On return to normal air, these mice develop retinal neovascularization, with peak development occurring 5 days after their return to normoxia. The pups received a 0.5-μL intravitreal injection of plasmid (2 mg/mL) OD on postnatal day one (P1). After the fifth day following return to normoxia (P17), the animals were euthanized, and the eyes removed and fixed in 4% paraformaldehyde and embedded in paraffin. Three hundred serial sections (6 μm) were cut sagittally through the cornea parallel to the optic disc. Every thirtieth section was placed on slides and stained with hematoxylin-eosin. This results in a total of 10 sections from each eye being scored in a masked fashion using light microscopy by counting endothelial cell nuclei extending beyond the inner limiting membrane into the vitreous.38 The efficacy of treatment with each plasmid was then calculated as the percent average nuclei per section in the injected eye versus the uninjected contralateral eye.
Time Course Analysis of Ribozymes
Time course experiments indicated that A2B receptor ribozymes rapidly cleaved synthetic targets either under standard in vitro conditions (20 mmol/L MgCl2) or at room temperature in physiological magnesium conditions (Figure 2). Figure 2A is an autoradiograph from a 10% polyacrylamide-8 mol/L urea gel used to separate the products of cleavage of the A2B Rz2 on the mouse A2B target for reactions performed at 37°C and in 20 mmol/L MgCl2. Figure 2B shows the graphical representation of the gel in Figure 2A and of the A2B Rz1 and the mouse target. This analysis was performed to determine the initial velocities for multiple-turnover reactions. This interval is typical when less than 15% of the substrate has been digested. The A2B Rz2 ribozyme exhibited a high rate of cleavage at 37°C in 20 mmol/L MgCl2 with 15% cleavage occurring at less than one minute. Because this interval was inconvenient for multiple-turnover kinetic analysis, we dropped the reaction temperature to 25°C in 1 mmol/L MgCl2 (Figure 2C).
Active A2B Rz2 Cleaves Both Mouse and Human Targets
The active A2B Rz2 ribozyme was designed to specifically cleave the mouse target sequence, but also cleaves the human target (Figure 3). As shown in Figure 3B the mouse target has a C at the 3′-end, and this will pair with a G in the ribozyme (Figure 1). The human target has a U at the 3′-end, and this will still form a non–Watson-Crick base pair with the G in the ribozyme and still permit cleavage.
Multiple-Turnover Kinetic Analysis of Ribozymes
Multiple-turnover kinetic analysis was performed on both A2B receptor ribozymes, and the kinetic parameters were determined (Table). Each analysis was performed a minimum of three times (P≤1×10−5). Reactions were done at both 37°C in 20 mmol/L MgCl2 (standard conditions) and at 25°C in 1 mmol/L MgCl2. The reactions at 37°C were terminated at 6 minutes for A2B Rz1 and one minute for A2B Rz2. The reactions at 25°C were terminated at 3 minutes for A2B Rz2. Based on Kcat, A2B Rz2 was 20-fold more active than A2B Rz1 under standard conditions, and retained good activity even in low magnesium and temperature. The A2B Rz2 was as active under these conditions as the naturally occurring hammerhead (from TRSV satellite RNA) under standard conditions.39 Because this A2B Rz2 was more potent than A2B Rz1, we used it exclusively for experiments in cell culture and in mice.
Effect of A2B Receptor Ribozymes on A2B Receptor mRNA Levels
The active A2B receptor ribozyme reduced A2B receptor mRNA levels in both HRECs and HEK293 cells to 70.5±5.4% (P=0.003) and 55.1±4.8% (P=5.1×10−5), respectively. Figure 4 shows the results of both relative quantitative RT-PCR and real-time RT-PCR analysis on mRNA isolated from HRECs and HEK293 cells transfected with the active and inactive versions of the A2B receptor ribozyme and with the control plasmid. As expected, the inactive A2B receptor ribozyme did not reduce mRNA levels.
Effect of A2B Receptor Ribozymes on Migration of HRECs
Plasmids expressing ribozymes or the cloning vector p21NewHp were transfected into HRECs to determine their effect on cell migration in a modified Boyden chamber. Cells transfected with plasmid coding for active A2B Rz2 reduced migration of cells to an average of 52.7±1.2% (P=7×10−4) when compared with the p21NewHp control at increasing concentrations of NECA (10 and 100 ng/mL), and cells transfected with plasmid coding for the inactive A2B Rz2 reduced migration of cells to an average of 86.8±9.0% (P=0.07) (Figure 5).
Effect of A2B Receptor Ribozymes on Neovascularization in the Neonatal Mouse Model of OIR
Plasmids expressing the ribozymes or the cloning vector p21NewHp were injected intravitreally on postnatal day one in the right eye of mouse pups, with no injection in the left eye. The pups and their dams were taken through the OIR.38 The extent of retinal angiogenesis was determined and the results are shown in Figure 6. Whereas p21NewHp had no effect on retinal angiogenesis, the active A2B Rz2 reduced the average number of nuclei per section on average by 53.5±8.2% (P=4.5×10−5). The inactive A2B Rz2 reduced the average number of nuclei per section on average by 10.5±7.4% (P=0.3). Figure 7 shows representative sections through a left-uninjected eye and a right eye injected with the active A2B Rz2 in the same mouse pup.
We have developed a hammerhead ribozyme that specifically cleaves the mouse and human adenosine A2B receptor mRNAs (Figures 2 and 3⇑). We have demonstrated that this ribozyme reduces the expression of the A2B receptor mRNA and the function of the A2B receptor in HEK293 cells and HRECs, respectively (Figures 4 and 5⇑). We found a 30% reduction in the A2B receptor mRNA signal in HRECs and a 45% reduction in the A2B receptor mRNA signal in HEK293 cells transfected with the active ribozyme compared with cells transfected with the control plasmid. The difference in the ability of the active ribozyme to reduce receptor mRNA levels between the two cell types most likely results from differences in transfection efficiency. This reduction in expression of the A2B receptor mRNA correlated with a reduction in A2B receptor function that we found in transfected HRECs. The chemotactic migration of HRECs across a porous membrane toward solutions containing increasing concentrations of NECA is dependent on the presence of A2B receptors on the cell surface.3,19 We found that the number of migrating cells diminished to 47% in cells expressing the active ribozyme relative to cells transfected with the empty vector. The reduction in migration of HRECs transfected with a plasmid expressing the active A2B receptor ribozyme suggests that cell surface levels of the A2B receptor protein have been reduced due to inhibition of expression of the A2B receptor mRNA by the ribozyme. These results provide strong evidence that cleavage of the A2B receptor mRNA reduces expression of the protein in cultured cells to a level that significantly inhibits the cellular function of the A2B receptor.
We also demonstrate that the active ribozyme inhibits preretinal neovascularization in vivo in a mouse model of OIR (Figure 6). Preretinal neovascularization was reduced 53% in eyes injected with the plasmid expressing the active ribozyme. These results suggest that this reduction in neovascularization is a result of ribozyme inhibition of the expression of the A2B receptor and demonstrate that this ribozyme could be useful for in vivo studies of A2B receptor function in retinopathies. By selectively reducing the expression of the A2B receptor mRNA, we are able to inhibit preretinal neovascularization in the mouse model of OIR and provide further evidence for the involvement of the A2B receptor in retinal angiogenesis.
One result of this study was to separate the inhibitory effects of the active ribozyme into its catalytic and antisense effects. Antisense DNA oligonucleotides have been shown to inhibit the physiological activity of the A2B receptor in cultured cells.29 We attribute the reduced migration after treatment with the catalytically inert inactive ribozyme to be caused by a minor antisense effect (Figure 5) rather than RNA reduction. Injection of the inactive ribozyme into the mouse model demonstrates that only a small portion of this reduction can be attributed to an antisense effect. Therefore, there is, at most, a minor antisense contribution in the reduction of preretinal neovascularization in the animal model. We conclude that ribozyme genes delivered as naked DNA lead to a substantial reduction in neovascularization in this model due to the catalytic activity of the ribozyme.
Finally, this study demonstrates the utility of using ribozymes to study complex physiological pathways. We successfully transfected human cells and an in vivo mouse model with a plasmid expressing a hammerhead ribozyme that reduced the expression of a single target mRNA. We were able to quantify the effects of the ribozyme through functional assays of the target. We will be able to use routine plasmid transfection of cells with vectors expressing the active ribozyme to study how inhibition of A2B receptor expression affects other components of this pathway. We also demonstrate that injection of the plasmid expressing the A2B receptor ribozyme into the mouse model reduces preretinal neovascularization. This model will allow us to further dissect the role of the A2B receptor in angiogenesis.
Grant support was provided by the Juvenile Diabetes Research Foundation International and NIH grants EY012601 and EY007739. The authors thank Tenning Maa (CVTherapeutics) for her technical assistance with real-time RT-PCR experiments.
D.Z. and L.B. have a commercial relationship with CVTherapeutics, Inc, Palo Alto, Calif.
Original received December 4, 2002; resubmission received June 19, 2003; revised resubmission received August 5, 2003; accepted August 5, 2003.
Dusseau J, Hutchins P. Hypoxia-induced angiogenesis in chick chorioallantoic membranes: a role for adenosine. Respir Physiol. 1988; 17: 33–44.
Grant MB, Tarnuzzer RW, Caballero S, Ozeck MJ, Davis MI, Spoerri PE, Feoktistov I, Biaggioni I, Shryock JC, Belardinelli L. Adenosine receptor activation induces vascular endothelial growth factor in human retinal endothelial cells. Circ Res. 1999; 85: 699–706.
Sexl V, Mancusi G, Holler C, Gloria Maercker E, Schutz W, Freissmuth M. Stimulation of the mitogen-activated protein kinase via the A2A-adenosine receptor in primary human endothelial cells. J Biol Chem. 1997; 272: 5792–5799.
Feoktistov I, Goldstein AE, Ryzhov S, Zeng D, Belardinelli L, Voyno-Yasenetskaya T, Biaggioni I. Differential expression of adenosine receptors in human endothelial cells: role of A2B receptors in angiogenic factor regulation. Circ Res. 2002; 90: 531–538.
Dusseau J, Hutchins P, Malbasa D. Stimulation of angiogenesis by adenosine on the chick chorioallantoic membrane. Circ Res. 1986; 59: 163–170.
Gidday JM, Park TS. Adenosine-mediated autoregulation of retinal arteriolar tone in the piglet. Invest Ophthalmol Vis Sci. 1993; 34: 2713–2719.
Gidday JM, Maceren RG, Shah AR, Meier JA, Zhu Y. KATP channels mediate adenosine-induced hyperemia in retina. Invest Ophthalmol Vis Sci. 1996; 37: 2624–2633.
Linden J, Thai T, Figler H, Jin X, Robeva AS. Characterization of human A2B adenosine receptors: radioligand binding, Western blotting, and coupling to Gq in human embryonic kidney 293 cells and HMC-1 mast cells. Mol Pharmacol. 1999; 56: 705–713.
Grant MB, Davis MI, Caballero S, Feoktistov I, Biaggioni I, Belardinelli L. Proliferation, migration, and ERK activation in human retinal endothelial cells through A2B adenosine receptor stimulation. Invest Ophthalmol Vis Sci. 2001; 42: 2068–2073.
Van Daele P, Van Coevorden A, Roger PP, Boeynaems JM. Effects of adenine nucleotides on the proliferation of aortic endothelial cells. Circ Res. 1992; 70: 82–90.
Takagi H, King G, Ferrara N, Aiello L. Hypoxia regulates vascular endothelial growth factor receptor KDR/Flk gene expression through adenosine A2 receptors in retinal capillary endothelial cells. Invest Ophthalmol Vis Sci. 1996; 37: 1311–1321.
Taomoto M, McLeod DS, Merges C, Lutty GA. Localization of adenosine A2a receptor in retinal development and oxygen-induced retinopathy. Invest Ophthalmol Vis Sci. 2000; 41: 230–243.
Mino RP, Spoerri PE, Caballero S, Player D, Belardinelli L, Biaggioni I, Grant MB. Adenosine receptor antagonists and retinal neovascularization in vivo. Invest Ophthalmol Vis Sci. 2001; 42: 3320–3324.
Dubey RK, Gillespie DG, Zacharia LC, Mi Z, Jackson EK. A2b receptors mediate the antimitogenic effects of adenosine in cardiac fibroblasts. Hypertension. 2001; 37: 716–721.
Dubey RK, Gillespie DG, Jackson EK. A2B adenosine receptors stimulate growth of porcine and rat arterial endothelial cells. Hypertension. 2002; 39: 530–535.
Shaw LC, Whalen PO, Drenser KA, Yan W, Hauswirth WW, Lewin AS. Ribozymes in treatment of inherited retinal disease. Methods Enzymol. 2003; 316: 761–776.
Agarwal A, Shiraishi F, Visner GA, Nick HS. Linoleyl hydroperoxide transcriptionally upregulates heme oxygenase-1 gene expression in human renal epithelial and aortic endothelial cells. J Am Soc Nephrol. 1998; 9: 1990–1997.
Smith LE, Wesolowski E, McLellan A, Kostyk SK, D’Amato R, Sullivan R, D’Amore PA. Oxygen-induced retinopathy in the mouse. Invest Ophthalmol Vis Sci. 1994; 35: 101–111.