Cloning of a chick A3 adenosine receptor: characterization of ligand binding and receptor-effector coupling of chick A1 and A3 adenosine receptors

Species dependence of A3 adenosine receptor pharmacology and function

Efforts to fully understand pharmacological differences between G protein-coupled receptor (GPCR) species homologues are generally not pursued in detail during the drug development process. To date, many GPCRs that have been successfully targeted are relatively well-conserved across species in amino acid sequence and display minimal variability of biological effects. However, the A3 adenosine receptor (AR), an exciting drug target for a multitude of diseases associated with tissue injury, ischemia, and inflammation, displays as little as 70% sequence identity among mammalian species (e.g., rodent vs. primate) commonly used in drug development. Consequently, the pharmacological properties of synthetic A3AR ligands vary widely, not only in binding affinity, selectivity, and signaling efficacy, but to the extent that some function as agonists in some species and antagonists in others. Numerous heterocyclic antagonists that have nM affinity at the human A3AR are inactive or weakly active at the rat and mouse A3ARs. Positive allosteric modulators, including the imidazo [4,5-c]quinolin-4-amine derivative LUF6000, are only active at human and some larger animal species that have been evaluated (rabbit and dog), but not rodents. A3AR agonists evoke systemic degranulation of rodent, but not human mast cells. The rat A3AR undergoes desensitization faster than the human A3AR, but the human homologue can be completely re-sensitized and recycled back to the cell surface. Thus, comprehensive pharmacological evaluation and awareness of potential A3AR species differences are critical in studies to further understand the basic biological functions of this unique AR subtype. Recombinant A3ARs from eight different species have been pharmacologically characterized thus far. In this review, we describe in detail current knowledge of species differences in genetic identity, G protein-coupling, receptor regulation, and both orthosteric and allosteric A3AR pharmacology.

Purines as potential morphogens during embryonic development

Components of purinergic signalling are expressed in the early embryo raising the possibility that ATP, ADP and adenosine may contribute to the mechanisms of embryonic development. We summarize the available data from four developmental models—mouse, chick, Xenopus and zebrafish. While there are some notable examples where purinergic signalling is indeed important during development, e.g. development of the eye in the frog, it is puzzling that deletion of single components of purinergic signalling often results in rather minor developmental phenotypes. We suggest that a key step in further analysis is to perform combinatorial alterations of expression of purinergic signalling components to uncover their roles in development. We introduce the concept that purinergic signalling could create novel morphogenetic fields to encode spatial location via the concentration of ATP, ADP and adenosine. We show that using minimal assumptions and the known properties of the ectonucleotidases, complex spatial patterns of ATP and adenosine can be set up. These patterns may provide a new way to assess the potential of purinergic signalling in developmental processes.

Adenosine induces growth-cone turning of sensory neurons

The formation of appropriate connections between neurons and their specific targets is an essential step during development and repair of the nervous system. Growth cones are located at the leading edges of the growing neurites and respond to environmental cues in order to be guided to their final targets. Directional information can be coded by concentration gradients of substrate-bound or diffusible-guidance molecules. Here we show that concentration gradients of adenosine stimulate growth cones of sensory neurons (dorsal root ganglia) from chicken embryos to turn towards the adenosine source. This response is mediated by adenosine receptors. The subsequent signal transduction process involves cAMP. It may be speculated that the in vivo function of this response is concerned with the formation or the repair and regeneration of the peripheral nervous system.

A1 adenosine receptor activation promotes angiogenesis and release of VEGF from monocytes

Adenosine is a proangiogenic purine nucleoside released from ischemic and hypoxic tissues. Of the 4 adenosine receptor (AR) subtypes (A1, A2A, A2B, and A3), the A2 and A3 have been previously linked to the modulation of angiogenesis. We used the chicken chorioallantoic membrane (CAM) model to determine whether A1 AR activation affects angiogenesis. We cloned and pharmacologically characterized chicken AR subtypes to evaluate the selectivity of various agonists and antagonists. Application of the A1 AR-selective agonist N6-cyclopentyladenosine (CPA; 100 nmol/L) to the CAM resulted in a 40% increase in blood vessel number (P<0.01), which was blocked by the A1 AR-selective antagonist C8-(N-methylisopropyl)-amino-N6-(5'-endohydroxy)-endonorbornan-2-yl-9-methyladenine (WRC-0571; 1 micromol/L). Selective A2A AR agonists did not stimulate angiogenesis in the CAM. In an ex vivo rat aortic ring model of angiogenesis that includes cocultured endothelial cells, fibroblasts, and smooth muscle cells, 50 nmol/L CPA did not directly stimulate capillary formation; however, medium from human mononuclear cells pretreated with CPA, but not vehicle, increased capillary formation by 48% (P<0.05). This effect was blocked by WRC-0571 (1.5 micromol/L) or anti-VEGF antibody (1 microg/mL). CPA (5 nmol/L) stimulated a 1.7-fold increase in VEGF release from the mononuclear cells. This is the first study to show that A1 AR activation induces angiogenesis. Stimulation of A2 ARs on endothelial cells results in proliferation and tube formation, and A2 and A3 ARs on inflammatory cells modulate release of angiogenic factors. We conclude that adenosine promotes a coordinated angiogenic response through its interactions with multiple receptors on multiple cell types.

Radioligand binding and functional responses of ligands for human recombinant adenosine A(3) receptors

The binding and functional properties of adenosine receptor ligands were compared in Chinese hamster ovary cells transfected with human adenosine A(3) receptors. Inhibition of [(125)I]-aminobenzyl-5'-N-methylcarboamidoadenosine ([(125)I]-AB-MECA) binding by adenosine receptor ligands was examined in membrane preparations. Inhibition of forskolin-induced cAMP accumulation by agonists was measured using a cAMP enzyme immunoassay. The rank order of agonist potency for both assays was N(6)-(3-iodobenzyl)-adenosine-5'-N-methyluronamide (IB-MECA) > 5'-N-ethylcarboxamidoadenosine (NECA) > (-)-N(6)-[(R)-phenylisopropyl] adenosine (R-PIA) > 4-aminobenzyl-5'-N-methylcarboxamidoadenosine (AB-MECA) > N(6)-cyclopentyl adenosine (CPA) > adenosine. The radioligand binding rank order of antagonist potency was N-[9-chloro-2-(2-furanyl)[1,2,4]-triazolo[1,5-c]quinazolin-5-benzeneacetamide (MRS1220) > 1,3-dipropyl-8-cyclopentylxanthine (DPCPX) > 8-phenyltheophylline (8-PT) > 8-(p-sulfophenyl)-theophylline (8-SPT). MRS1220 competitively inhibited the effect of IB-MECA on cAMP production, with a K(B) value of 0.35 nm. These data are characteristic of adenosine A(3) receptors. The absence of Mg(2+) and presence of guanosine 5'-(gamma-thio)triphosphate (GTPgammaS) significantly reduced agonist binding inhibition potency, indicating binding to high- and low-affinity states. The IB-MECA, NECA and R-PIA IC(50) values were greater for the cAMP assay than for radioligand binding, suggesting an efficient stimulus-response transduction pathway.

Adenosine signaling promotes neuronal, catecholaminergic differentiation of primary neural crest cells and CNS-derived CAD cells

In neural crest (NC) cultures cAMP signaling is an instructive signal in catecholaminergic, sympathoadrenal cell development. However, the extracellular signals activating the cAMP pathway during NC cell development have not been identified. We demonstrate that in avian NC cultures, evidenced by tyrosine hydroxylase expression and catecholamine biosynthesis, adenosine and not adrenergic signaling, together with BMP2, promotes sympathoadrenal cell development. In NC cultures, addition of the adenosine receptor agonist NECA in the presence of BMP2 promotes sympathoadrenal cell development, whereas the antagonist CGS 15943 or the adenosine degrading enzyme adenosine deaminase (ADA) suppresses TH expression. Importantly, NC cells express A2A and A2B receptors which couple with Gsalpha increasing intracellular cAMP. Employing the CNS-derived catecholaminergic CAD cell line, we also demonstrate that neuronal differentiation mediated by serum withdrawal is further enhanced by treatment with IBMX, a cAMP-elevating agent, or the adenosine receptor agonist NECA, acting via cAMP. By contrast, the adenosine receptor antagonist CGS 15943 or the adenosine degrading enzyme ADA inhibits CAD cell neuronal differentiation mediated by serum withdrawal. These results support that adenosine is a physiological signal in neuronal differentiation of the CNS-derived catecholaminergic CAD cell line and suggest that adenosine signaling is involved in NC cell development in vivo.

Interactions of Nucleoside Analogs, Caffeine, and Nicotine with Human Concentrative Nucleoside Transporters 1 and 2 Stably Produced in a Transport-Defective Human Cell Line

Pharmacologically important drugs were examined as potential inhibitors or permeants of human concentrative nucleoside transporters 1 (hCNT1)- and 2 (hCNT2)-producing stable transfectants by assessing their abilities to inhibit uridine transport. hCNT1 exhibited high affinities for uridine analogs (5-fluorouridine, 2'-deoxyuridine, 5-fluoro-2'-deoxyuridine, and 5-fluoro-5'-deoxyuridine) with K(i) values of 22 to 33 microM, whereas hCNT2 exhibited moderate affinities for 5-fluoro-2'-deoxyuridine, high affinities for 2'-deoxyuridine and 5-fluorouridine, and low affinity for 5-fluoro-5'-deoxyuridine. The uridine analogs were transported at 2-fold higher rates (at 10 microM) by hCNT1 than by hCNT2. Enantiomeric configuration and the 3'-hydroxyl group of the ribose ring were important determinants for interaction with hCNTs, whereas the 2'-hydroxyl group was less important. Both transporters bound N(6)-(p-aminobenzyl)adenosine with affinities similar to those of adenosine (K(i) = 28-39 microM). Other adenosine receptor ligands, including caffeine, bound better to hCNT1 than to hCNT2 (K(i) = 46 versus 103 microM, respectively), whereas 2-chloroadenosine bound better to hCNT2 than to hCNT1 (K(i) = 37 and 101 microM, respectively). There was a greater than 3-fold difference in binding affinities between hCNT1 and hCNT2 for nicotine (K(i) = 63 versus 227 microM). However, direct measurements of nicotine and caffeine uptake rates (10 microM) failed to demonstrate mediated uptake by either transporter. Although hCNT1 bound several adenosine analogs relatively well, it did not transport 2-chloro-2'-deoxyadenosine (cladribine) or 2-fluoro-9-beta-d-arabinofuranosyladenine (fludarabine), whereas hCNT2 transported both, albeit with low activities. The results indicated that although hCNT1 and hCNT2 possess some overlap in transport of several uridine and adenosine analogs, they also exhibit distinct differences in capacity to interact with some adenosine receptor ligands, adenosine-based drugs, and nicotine.

Cyclic GMP-Dependent Protein Kinases and the Cardiovascular System

Signaling cascades initiated by nitric oxide (NO) and natriuretic peptides (NPs) play an important role in the maintenance of cardiovascular homeostasis. It is currently accepted that many effects of these endogenous signaling molecules are mediated via stimulation of guanylyl cyclases and intracellular production of the second messenger cGMP. Indeed, cGMP-elevating drugs like glyceryl trinitrate have been used for more than 100 years to treat cardiovascular diseases. However, the molecular mechanisms of NO/NP signaling downstream of cGMP are not completely understood. Recent in vitro and in vivo evidence identifies cGMP-dependent protein kinases (cGKs) as major mediators of cGMP signaling in the cardiovascular system. In particular, the analysis of conventional and conditional knockout mice indicates that cGKs are critically involved in regulating vascular remodeling and thrombosis. Thus, cGKs may represent novel drug targets for the treatment of human cardiovascular disorders.

Adenosine and opioid receptor-mediated cardioprotection in the rat: evidence for cross-talk between receptors

The relative roles of free-radical production, mitochondrial ATP-sensitive K+ (mitoKATP) channels and possible receptor cross-talk in both opioid and adenosine A1 receptor (A1AR) mediated protection were assessed in a rat model of myocardial infarction. Sprague-Dawley rats were subjected to 30 min of occlusion and 90 min of reperfusion. The untreated rats exhibited an infarct of 58.8 +/- 2.9% [infarct size (IS)/area at risk (AAR), %] at the end of reperfusion. Pretreatment with either the nonselective opioid receptor agonist morphine or the selective A1AR agonist 2-chloro-cyclopentyladenosine (CCPA) dramatically reduced IS/AAR to 41.1 +/- 2.2% and 37.9 +/- 5.5%, respectively (P < 0.05). Protection afforded by either morphine or CCPA was abolished by the reactive oxygen species scavenger N-(2-mercaptopropionyl)glycine or the mitoKATP channel blocker 5-hydroxydecanoate. Both morphine- and CCPA-mediated protection were attenuated by the selective A1AR antagonist 1,3-dipropyl-8-cyclopentylxanthine and the selective delta1-opioid receptor (DOR) antagonist 7-benzylidenealtrexone. Simultaneous administration of morphine and CCPA failed to enhance the infarct-sparing effect of either agonist alone. These data suggest that both DOR and A1AR-mediated cardioprotection are mitoKATP and reactive oxygen species dependent. Furthermore, these data suggest that there are converging pathways and/or receptor cross-talk between A1AR- and DOR-mediated cardioprotection.