Purinergic Signalling

, Volume 2, Issue 1, pp 1–324

Invited Lectures

Overviews Purinergic signalling: past, present and future
Open Access

Overviews Purinergic signalling: past, present and future

Professor Geoffrey Burnstock PhD DSc FAA FRCS(Hon) FRCP(Hon) FMedSci FRS President, Autonomic Neuroscience Centre

Following a brief account of the early history of the discovery of purinergic signalling, a personal view of some of the exciting cutting-edge directions being taken by research in the field will be considered. In particular, emphasis will be placed on the pathophysiology of purinergic signalling and its therapeutic potential.

A Fuller Quiver Adenosine Receptors as Therapeutic Targets

Bruce N. Cronstein, MD

NYU School of Medicine, 50 First Ave., New York, NY 10016 Cronsb01@med.nyu.edu

Recent advances in understanding the role of adenosine and its receptors in physiology and pathophysiology as well as new developments in medicinal chemistry of these receptors have made it possible to begin to realize the therapeutic potential of adenosine and its receptors. Currently adenosine itself is targeted at the heart; a bolus of adenosine intravenously is commonly used to treat supraventricular tachycardia whereas an infusion of adenosine is used as a coronary vasodilator during pharmacologic stress testing. Non-selective adenosine receptor antagonists are used to maintain wakefulness (caffeine), as an analgesic (caffeine) and, less commonly at present, to treat bronchospasm (theophylline, aminophylline, enprofylline). Currently a number of new selective adenosine receptor agonists and antagonists are in testing for a variety of new indications and one adenosinetargeted indication. The older indication is pharmacologic stress testing and new selective A2A receptor agonists, ATL146e (Apadenoson, Adenosine Therapeutics) and Regadenoson (CV Therapeutics), are currently under study (Phase II–III) for this indication. In addition, a selective A1 receptor agonist is in trials for supraventricular tachycardia (tecadenoson, CV Therapeutics), Currently a number of companies are testing selective adenosine receptor agonists for newer indications including Rheumatoid Arthritis, cancer and wound healing, Adenosine A3 receptor agonists are in trials for the treatment of Rheumatoid Arthritis and Cancer (CF101, CF102, Can-Fite Biopharmaceuticals) and preliminary promising results in patients with RA have been reported. Topical application of an adenosine A2A receptor agonist to promote healing of diabetic foot ulcers is currently in Phase II trials after the successful completion of a Phase I safety/early efficacy trial (MRE94, King Pharmaceuticals). An allosteric enhancer at the adenosine A1 receptor is in trials for neuropathic pain (T-62, King Pharmaceuticals). Because adenosine receptor stimulation may also be involved in disease pathogenesis selective adenosine receptor antagonists are being studied for select indications as well. Istradefylline (KW-6002, Kyowa Pharmaceuticals), a selective adenosine A2A receptor antagonist, is now in late stage clinical trials for the treatment of Parkinson's Disease. Another approach to targeting adenosine receptors is to increase extracellular adenosine concentrations, mechanisms shown to mediate the anti-inflammatory effects of two drugs commonly used in the treatment of Rheumatoid Arthritis, methotrexate and sulfasalazine. Dipyridamole inhibits adenosine uptake and this clearly underlies its effect as a coronary vasodilator during pharmacologic cardiac stressing. New methotrexate analogues are in therapeutic trials in Rheumatoid Arthritis and recent pre-clinical studies of an adenosine uptake inhibitor for the treatment of inflammatory arthritis have been reported (KF24345, Kyowa Hakka Kogyo). In the future other therapeutic indications for adenosine receptor targeting may include fibrosis and scarring in the skin or liver, inflammation and inflammatory diseases, chronic pain syndromes and other yet to be described conditions.

Activation of Adenosine A2A Receptors on CD4+ T Cells Reduces Reperfusion Injury

Joel Linden

Department of Medicine, Cardiovascular Research Center, University of Virginia, Charlottesville, VA, 22908 USA

Agonists of adenosine A2A receptors (A2AR) such as ATL146e can substantially reduce necrosis of liver, heart and other tissues when added at the time of reperfusion following ischemia. Protection is attenuated by A2AR antagonists and is absent in A2AR KO mice. We prepared bone marrow chimera mice by transplanting bone marrow from A2AR KO mice to previously irradiated WT syngenic recipients (KO/WT chimera) or vice versa. Experiments with these mouse chimera revealed that A2AR-mediated protection from liver or heart reperfusion injury is entirely dependent on A2ARs on bone marrow-derived cells. A mouse line with a floxed adora2a gene was constructed to enable specific deletion of receptors in tissues expressing Cre recombinase. The lysM promoter was used to express Cre selectively in macrophages and neutrophils. LysMCre-A2ARf/f mice effectively delete the A2AR gene, adora2a, in macrophages and neutrophils but not T cells based both on functional assays and PCR based assays to quantify the floxed and recombined genes in purified cell populations. ATL146e was able to inhibit reperfusion injury in LysMCre-A2ARf/f mice, indicating that bone marrow derived cells other than neutrophils and macrophages contribute to protection. Macrophages participate in liver reperfusion injury because injury is reduced by depletion of macrophages with liposomal clodronate. Lymphocytes also appear to participate in reperfusion injury because Rag1 KO mice that lack lymphocytes and B cells are protected. Reperfusion injury is restored in Rag1 KO mice by adoptive transfer of CD4+, but not CD8+ T cells, and injury is not restored by CD4+ cells derived from INFγ KO mice. Assays of CD4+ T cells in vitro indicates that activation of the TCR causes rapid induction of A2AR mRNA, and that ATL146e binding to the A2AR can reduce TCRstimulated INFγ production by 98%. ATL146e reduces reperfusion injury in Rag1 KO mice following adoptive transfer of WT, but not A2AR KO CD4+ cells purified from donor spleens. As confirmation that CD4+ cells are important in reperfusion injury we found that depletion of these cells with anti-CD4+ antibodies also reduced liver or hear reperfusion injury but anti-CD8+ antibodies that effectively deplete CD8+ cells from blood are ineffective. Immunohistochemistry was used to measure the time course of accumulation of T cells into the heart and liver after initiating reperfusion following ischemia. In both tissues there is a small but significant accumulation of T cells within minutes that is largely attenuated by ATL146e. These data suggest that an important means by which A2AR activation reduces reperfusion injury is by activating receptors on CD4+ T lymphocytes. This reduces macrophage-dependent T cell activation and INFγ release required for further rapid recruitment of additional lymphocytes and/or subsequent recruitment of neutrophils into the tissue. In ongoing work we are attempting to determine what factors are responsible for activating CD4+ T cells during reperfusion injury and if a particular subpopulation of CD4+ T cells are selectively activated. A2A agonists may be clinically useful for the treatment of ischemia-reperfusion injury, e.g. during coronary reperfusion following myocardial infarction, or tissue transplantation.

Berne Award Lecture: Targeting Adenosine Receptors

Bertil B. Fredholm

Department of Physiology and Pharmacology, Karolinska Institutet, S-171 77 Stockholm, Sweden Bertil.fredholm@ki.se

My work in the adenosine area started, inspired by the professors Berne, Gerlach and Fain, with attempts to examine the physiological roles of adenosine in circulation and metabolism. It was realised early—especially by Rall and Daly—that methylxanthines including theophylline and caffeine were very useful tools to examine roles of adenosine, because they acted as receptor antagonists. The cardiovascular, central nervous and metabolic actions of adenosine also prompted several investigators, not least Westermann, to initiate development of adenosine analogues with some level of selectivity. Using these as tools we discovered that adenosine and ATP reduced neurotransmitter release, something that was independently discovered by Ribeiro.

The work of Van Calker, Hamprecht and Londos had clarified that there are at least two receptor subtypes, which we now call A1 and A2. With many collaborators, especially the late Tom Dunwiddie, we explored the roles of A1 and A2 receptors in brain tissue and related these actions to studies of adenosine formation. To the latter end, we used and modified techniques pioneered by McIlwain and Daly, and found with Ungerstedt (in the first ever study using microdialysis) that there are resting levels of adenosine that are substantial (20–300 nM). The effects of adenosine on transmitter release and neurotransmission were mediated by A1 receptors, whereas the marked effect on lymphocytes and blood vessels were mediated by A2 receptors. Radioligands were developed by several different groups and this enabled careful studies of receptor distribution. Autoradiographic techniques were especially important. When adenosine receptors were cloned this allowed even more precise localization studies and it was found that adenosine A2A receptors are co-localized with dopamine D2 receptors in one cell type of the basal ganglia and related structures. The A2A receptors were found to be of tremendous importance in the basic physiology of the basal ganglia and to be involved in many drug actions and physiological responses.

Cloning of the adenosine receptors also allowed the generation of knock-out mice to pinpoint the roles of the different receptors. These mice have shown that adenosine is not critically important for development or basic physiology, and that adenosine has, as suspected from the earlier investigations, a predominantly modulatory role. Therefore modulatory adaptations are limited and adenosine receptors are interesting drug targets. Using A2Areceptor knock-out mice it has been shown that these receptors are of fundamental importance for the stimulatory actions of caffeine and for its effects on arousal. It is also shown that A2A receptors can be a target for drugs to combat Parkinsońs disease, as well as ischemia. Mice with targeted deletions of A1receptors have shown that these receptors are critically important in kidney physiology. A1 receptors are important in limiting epileptic seizures and resulting neuronal death. They are very important in regulating transmitter release, but apparently play little role in ischemic cell death. This suggests that excitatory amino acid release may be overemphasized as a mechanism of neuronal cell death after ischemia. By contrast, A1 receptors are very important in mediating protective effects in the heart, cause analgesia, and they play a role in metabolic adaptations.

Thus, many years of research targeting adenosine receptors have shown that they are important in some, but not many, physiological process. Adenosine receptors, do, however play several important pathophysiological roles. Therefore drugs that target adenosine receptors have a future.

P2X and P2Y receptor-involvement in pain sensation; possible targets for new analgesics

Peter Illes1, Wolfgang Schröder2 and Zoltan Gerevich1

1Rudolf-Boehm-Institute for Pharmacology and Toxicology, University of Leipzig, D-04107 Leipzig and 2Department of Pharmacology, Grünenthal GmbH, D-52099 Aachen, Germany illp@medizin.uni-leipzig.de

Cultured dorsal root ganglion (DRG) neurons of rats are known to be endowed at their cell bodies as well as at their peripheral and central processes with P2X3 receptors mediating rapidly desensitizing inward current responses to ATP or its structure analogue α,β-methylene ATP (α,β-meATP). In addition, we described the presence of P2Y1 receptors at these neurons, negatively coupled to N-type voltage-sensitive Ca2+ channels. Whereas P2X3 receptor-activation is supposed to induce propagated action potentials and thereby pain sensation, P2Y receptor-activation may cause an opposite effect, by decreasing the Ca2+ dependent release of sensory transmitters from the central terminals of DRG neurons. According to the known transduction mechanisms of P2X3 and P2Y1 receptors an early and rapidly declining pronociceptive action of ATP is expected to be followed by a more protracted antinociceptive action of the same compound. In support of this assumption, we found a negative interaction between P2Y1 and P2X3 receptors at rat DRG neurons, which was due to G protein activation, but did not utilize any of the hitherto known second-messenger mechanisms including phospholipase C, phosphatidylinositol 3 kinase, protein kinase C, inositoltriphosphate or Ca2+/calmodulin kinase II. In accordance with our in vitro data the intrathecal application of P2Y receptor agonists exhibited analgesic activity in an acute pain model (tail-flick test). In conclusion, the endogenous P2X/P2Y receptor agonist ATP either by itself or via its degradation product ADP may on the one hand induce pain and on the other hand may limit an overt pain reaction. These two opposing mechanisms may operate with a considerable time-lag between each other.

Structure and function in P2X receptors

R. Alan North

Faculty of Life Sciences, University of Manchester, Manchester, M13 9PT, UK

P2X receptors belong to a superfamily of membrane proteins that are characterised by two membrane-spanning domains and a cysteine-rich ectodomain. Other members of this superfamily include (a) the nucleoside triphosphate diphosphohydrolase enzymes (NTPDase 1–8) which catalyse the extracellular degradation of ATP and ADP, and (b) the family of ion channels that includes the acid-sensing ion channels (ASIC 1–4), epithelial sodium channels (ENaC α–δ) and degenerins of C. elegans (e.g. MEC-4, MEC-10, DEG-1, DEL-1, UNC-8, UNC-105). The three families of proteins are unrelated in primary structure. In terms of quaternary structure, P2X receptors are considered to be trimers, whereas the body of evidence suggest that ENaC/ASIC channels are tetramers. This lecture will summarise and review experiments in which site-directed mutagenesis and functional expression have been used to deduce those parts of the P2X receptor involved in (a) ATP binding, (b) ion permeation, and (c) interactions with other proteins. The principal focus will be on P2X1, P2X2, P2X2/3 and P2X4 subunits, with comparisons from studies on other family members where appropriate.

This work was supported by Wellcome Trust.

The ecto-nucleotidase CD39/NTPDase1 is a key modulator of vascular inflammation and immunity

Simon C. Robson

Liver and Transplant Centers, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston. MA, USA srobson@bidmc.harvard.edu

Extracellular nucleotides (e.g. ATP, ADP, UTP) activate type-2 purinergic/pyrimidinergic (P2Y and P2X) receptors on platelets, endothelium and leukocytes. Ecto-nucleotidases hydrolyze these mediators, ultimately to the respective nucleosides, to regulate P2-signaling. Ecto-nucleotidases of the CD39/E-NTPDase family are expressed at high levels in the vasculature and immune systems. In addition to catalytic functions of the ectodomain of CD39, the palmitoylated intracytoplasmic N-terminus has been shown to structurally and functionally associate with a Ran binding protein, termed RanBPM. This multi-adaptor, scaffolding membrane protein regulates small GTPases and influences integrin signaling.

We have proposed that spatial and temporal expression of CD39/NTPDase1 within the vasculature, by immune cells and/or derived microparticles (viz. membrane vesicles) could regulate inflammatory processes, immune reactions and also impact development of cancers.

Expression of vascular CD39 appears crucial in regulating innate immunity, platelet thrombotic reactions, acute ischemic insults, altered vascular permeability, angiogenesis and tumor growth. For example, as visualized by in vivo video-microscopy, laser-induced arteriolar thrombus is characterized by rapid accumulation of platelets and microparticles. This process is then stabilized by platelet disaggregation with decreases in thrombus mass. The accumulation of NTPDase1 within thrombi blocks further ADP-mediated platelet activation. Mutant mice null for Cd39 and transgenic over-expressors of CD39 show the predicted abnormalities with marked differences in clot formation in vivo.

Dendritic cell expression of CD39 influences antigen presentation and T cellular responses that are crucial in the evolution of adaptive immunity e.g. alloimmune reactions. CD39 is both an important surface marker of T regulatory cells (Treg) and an integral functional component of these cells. Co-ordinated expression of CD39 on Treg and the adenosine A2A receptor on activated effector T cells (Teff) generates a paracrine, immunosuppressive loop. Adoptive transfer of Cd39 null Treg fails to inhibit allograft rejection in vivo. Furthermore, Cd39 null mice develop autoimmune manifestations with deviated Th1 responses.

In addition to major recognized thromboregulatory roles, CD39 expression also has functional relevance for cellular immunoregulation, in both allo- and autoimmune reactions. These findings suggest integration of vascular inflammatory and immunologic purinergic mechanisms. Pharmacologic modalities to modulate or boost NTPDase1 expression may suppress unwanted, deleterious vascular or immune reactions, as seen in autoimmune disease and transplant graft rejection. In turn, related approaches could be employed to augment host protective responses promoting tissue regeneration and normal repair processes.

Grant support from the National Institutes of Health (HL57307, HL63972 and HL076540).

Giuliana Fassina Award: Therapeutic Potential of Partial A1Agonists in Insulin Resistance and Diabetes

Luiz Belardinelli, John Shryock, Arvinder Dhalla

Department of Pharmacological Sciences, CV Therapeutics Inc. Palo Alto, CA. USA 94304 luiz.belardinelli@cvt.com

A1 adenosine receptor (A1AdoR) agonists are potent anti-lipolytic agents that inhibit adipose tissue lipolysis and lower circulating free fatty acids (FFA) levels. A reduction of lipolysis in adipocytes is of potential benefit in treatments of dyslipidemia, type II diabetes, and metabolic syndrome. Therefore, an A1AdoR agonist that reduces lipolysis in adipocytes may be useful in the therapy of insulin-resistant states. However, A1AdoR agonists have potential unintended side effects as a result of the presence of A1AdoR in many tissues in addition to the adipose tissue. Functional selectivity of drug action (maximal or near-maximal anti-lipolytic effect with minimal or no cardiovascular effects) can be achieved by exploiting the differential receptor-effector coupling between adipose tissue and cardiac tissue. The undesired effects of A1AdoR in non-adipose tissues can be further minimized by use of low-efficacy agonists or partial agonists. We have discovered a novel partial agonist (CVT-3619) of the A1AdoR that inhibits adipose tissue lipolysis and lowers circulating FFA levels. CVT-3619 is more than 10–30-fold selective for A1 vs other AdoRs as determined by binding assays. CVT-3619 reduces cyclic AMP content and release of FFA from rat epididymal adipocytes with IC50 values (potency) of 6 and 44 nM, respectively. CVT-3619 is a partial agonist relative to CPA (a full agonist) to reduce lipolysis in rat isolated adipocytes. CVT-3619 (at concentrations of 10 nM-30 µM) does not change atrial rate in rat isolated heart and causes a small (6-msec) prolongation of the AV nodal conduction time without causing atrioventricular block in guinea pig isolated heart. In in-vivo studies, CVT-3619 (1–10 mg/kg) decreases both FFA and triglyceride (TG) levels (15–57%) in a dose-dependent manner in awake rats. CVT-3619 does not have any significant effect on heart rate and blood pressure at the doses that have maximal anti-lipolytic effects. At higher doses of CVT-3619 there is a significant but small decrease in heart rate. CVT-3619 also increases the potency of insulin to decrease lipolysis by 3-fold, suggesting that CVT-3619 increases insulin sensitivity. The potential anti-diabetic effect of CVT-3619 (10 mg/kg sc given twice daily for 5 consecutive days) was studied in ZDF rats, an animal model of type II diabetes. CVT-3619 treatment significantly lowers fasting plasma glucose, insulin, FFA and triglycerides as compared to vehicle treated rats. An oral glucose tolerance test performed on the 5th day of treatment with CVT-3619 shows an improvement in insulin sensitivity as determined by the area under the curve of the time-dependent changes in plasma glucose and insulin levels (AUCg × AUCi). In conclusion, CVT-3619 is a partial, orally bioavailable A1AdoR agonist that lowers circulating FFA and TG levels resulting in improved insulin sensitivity, with minimal cardiovascular effects.

Burnstock Lecture: Versatile messengers in the brain: Nucleotide storage, signalling and hydrolysis

H. Zimmermann

Institute of Cell Biology and Neuroscience, Biocenter, J.W. Goethe-University, 60439 Frankfurt am Main Germany h.zimmermann@cns.uni.frankfurt.de

ATP was first identified in muscle extracts in 1929. Proof for its role as an extracellular signaling molecule became apparent considerably later and was not without controversy in the beginning. Some of the strongest support for a potential extracellular function of ATP came from the demonstration of its storage inside granules of a variety of secretory cells such as in blood platelets and in chromaffin granules in the 1950ies, and later in adrenergic (1958/1963) and cholinergic (1974) synaptic vesicles. ATP was found to be costored with the noradrenaline or acetylcholine. On nerve stimulation, ATP is depleted from cholinergic synaptic vesicles in parallel with acetylcholine and is replenished together with acetylcholine during a subsequent period of rest. Adenosine taken up into cholinergic nerve terminals via a high affinity transporter becomes immediately phosphorylated, ends up in synaptic vesicles in the form of ATP and becomes coreleased with acetylcholine (Zimmermann, 1982) [1]. The costrorage of ATP with neurotransmitter substances as well as the coexistence of neuropeptides with classical neurotransmitters raised the question whether neurons may be able to employ more than one signaling substance, a new principle borne out by Geoffrey Burnstock in 1976 [2].

Also the notion of cell surface-located ATPases reaches far back into the 1950ies, based on both biochemical and enzyme histochemical evidence. Yet the functional role of extracellular ATP hydrolysis remained a matter of speculation. The evidence for an involvement of extracellular nucleotides in a variety of physiological functions increased dramatically form the 1970ies onwards. But the general acceptance of nucleotide signaling by the scientific community began only with the molecular cloning, heterologous expression and functional characterization of the molecular players involved. The first one was ecto-50-nucleotidase (1990) followed by P2 receptors (1993), and NTPDase1 (CD39) (1996).

By now, nucleotides have been demonstrated to control major functions in the nervous system. These include synaptic transmission and its modulation, reciprocal signaling between neurons and glia, propagation of glial calcium waves, induction of gliosis or also activation, modulation and transmission in sensory systems. A major role of ecto-nucleotidases in these settings is in the modulation of ligand availability at nucleotide and nucleoside receptors. A novel and interesting function of nucleotides uncovered more recently concerns their role in adult and embryonic neurogenesis where ATP and UTP cooperate with growth factor receptors in activating progenitors (Mishra et al., 2006) [3]. This and other recent examples highlight the functional importance of interactive pathways between signaling via nucleotides and other cellular messengers.

Basic Pharmacology

Molecular Physiology and Pharmacology of P2X1 Receptors

Jonathan Roberts, Catherine Vial, Helen Digby, Kelvin Agboh, Hairuo Wen & Richard J. Evans

Department of Cell Physiology and Pharmacology, Medical Sciences Building, University of Leicester, Leicester, U.K. LE1 9HN

ATP binds to the large extracellular loop of P2X receptors and results in channel activation. It seems likely that residues important for ATP action at the receptor would be conserved in the family and in the extracellular loop 6 Invited Lectures of the P2X receptors >90 of the ∼280 amino acids are identical in at least 5 of the seven P2X receptor subunits. Alanine replacement mutagenesis of the majority of conserved amino acids in the extracellular loop of the P2X1 receptor has little or no effect on the response to ATP indicating that they do not play an essential role. The results from systematic alanine scanning of conserved residues in the extracellular loop give rise to a model of the site of ATP action at the P2X1 receptor with K68 and K309 binding to the phosphate tail and the motifs F185T186 and N290F291R292 co-ordinating the binding of the adenine ring. A similar binding environment has been described for the crystal structure of rat synapsin II. To test and refine the model we have used cysteine scanning mutagenesis of the region S286–I329 to investigate the environment adjacent to important conserved amino acids N290F291R292 and K309 and the residues close to the second transmembrane domain. Methanthiosulphonate (MTS) compounds can be used to modify any accessible free cysteine residues. For the majority of cysteine mutants there was little or no effect on ATP evoked responses or modification by either positively charged MTSEA or negatively charged MTSES. At mutants D316C, G321C, A323C and I328C responses were sensitive to MTS compounds, e.g. an ∼50% reduction in response to ATP on application of MTSEA for G321C. However at these mutants there was no shift in ATP potency of the point mutants from WT either before or after MTS reagent application. These results indicate that the modification of ATP response results from an effect on ionic permeation and not on agonist binding. At mutants N290C, F291C, R292C and K309C responses were sensitive to MTS reagent application, however in these cases ATP potency was significantly reduced compared to WT channels and MTS reagents resulted in a change in ATP potency, for example at K309C the EC50 was decreased ∼40 fold on addition of positively charged MTSEA showing that charge at this position is important. These support previous findings and suggest that these residues mediate part of the ATP binding pocket.

Supported by the Wellcome Trust.

New insights into P2X7 receptor signaling

Annmarie Surprenant

Institute of Molecular Physiology, Department of Biomedical Science, University of Sheffield, Sheffield, UK a.supprenant@sheffield.ac.uk

Abstract not received

Novel Signal Transduction Pathways Regulated by P2Y2 Nucleotide Receptors Mediate Inflammatory Responses in Mammalian Cells

Weisman, G.A., Seye, C.I., Liao, Z., Yu, N., Wang, M., Liu, J., Chorna, N.,1 Baker, O., Camden, J., Sun, G.Y., Gonzalez, F.A.,1 and Erb, L.

University of Missouri-Columbia, Columbia, MO 65212 USA and *University of Puerto Rico, San Juan, PR 00931 USA weismang@missouri.edu

The Gq/11-coupled P2Y2 nucleotide receptor (P2Y2R) for ATP and UTP activates phospholipase C leading to an increase in IP3-dependent calcium mobilization and diacylglycerol-dependent activation of protein kinase C, responses that regulate the activity of phospholipases A2 in primary murine astrocytes. Recent studies indicate that P2Y2Rs possess novel molecular determinants that enable them to activate other signaling pathways independent of Gq proteins. For example, data indicate that proline rich Src-homology-3 (SH3) binding domains in the intracellular C-terminus of the P2Y2R can mediate the transactivation of growth factor receptors (e.g., vascular endothelial growth factor receptor-2) to promote the up-regulation of cell adhesion molecules (e.g., vascular cell adhesion molecule-1) in endothelium that promote the binding and vascular infiltration of monocytes associated with inflammatory responses in cardiovascular diseases. In addition, an arginine-glycine-aspartic acid (RGD) motif in the extracellular domain of the P2Y2R mediates interactions with αvβ35 integrins that are required for Go protein activation and nucleotide-induced increases in the motility of primary astrocytes, and G12-mediated stress fiber formation that regulates cytoskeletal rearrangements required for cell chemotaxis. The P2Y2R is also capable of activating matrix metalloproteases including the adamalysins ADAM 10 and ADAM 17 that increase the degradation of amyloid precursor protein in astrocytoma cells to generate the non-amyloidogenic peptide s-APPα, suggesting that P2Y2R function may be neuroprotective in Alzheimer's disease. Other data indicate that P2Y2R-mediated activation of metalloproteases in salivary gland epithelial cells plays a role in the up-regulation of cell adhesion molecules that facilitate lymphocyte binding, a response associated with the auto-immune disease Sjögren's syndrome. Taken together, our results indicate that P2Y2R expression in a variety of mammalian cell types can promote inflammatory responses due to the ability of the P2Y2R to activate diverse signaling pathways in addition to the well-established activation of Gq proteins. This study was supported by the National Institutes of Health Grants 1 P01-AG-018357 and 1 R01-DE-07389.

Nucleotide presynaptic ionotropic receptors are accurately tuned by other neurotransmitters

Miguel Díaz-Hernández, Rosa Gomez-Villafuertes, Javier Gualix, Maria Diez-Zaera, Miriam León-Otegui and M. Teresa Miras-Portugal

Presynaptic functional P2X receptors and dinucleotide receptors are widely present in the CNS. Calcium responses induced by purinergic agonists evoke neurotramitter stores release of synaptic vesicles from Cholinergic, GABAergic and Glutamatergic nerve terminals. However, when a single cholinergic terminal was stimulated with nucleotide and nicotinic agonists altogether results in a significant decrease of the [Ca2+]i signalling compared to responses of each independent agonist. Inhibitory interaction between both receptors is reverted when one of them is blocked by specific antagonists. The receptor's inhibitory cross-talk confirms the involvement of calcium/calmoduline-dependent protein kinase II, CaMKII, as the inhibitory effects are reverted in the presence of the specific inhibitors KN-62 and KN-93.

In GABAergic nerve endings it has been described that activation of GABA-B autoreceptors positively modulated the ATP and dinucleotide responses, inducing a significantly increase on the affinity of purinergic receptors.

Finally preliminary results obtained on Glutamatergic synaptic terminals appear indicate the existence of an inhibitory interaction between P2X and glutamatergic autoreceptors.

All these results demonstrate the existence of an efficient interaction between purinergic and other autoreceptors populations, opening a new understanding on the functional regulation of terminals releasing the most abundant neurotransmitters in the CNS.

Quantification and agonist-promoted regulation of the P2Y1 receptor

T. K. Harden, D. Houston, D.M. Bourdon, G.L. Waldo, M. Ohno, and K.A. Jacobson

Dept. Pharmacology, University of North Carolina School of Medicine, Chapel Hill, NC, USA and NIDDKD, NIH, Bethesda, MD, USA tkh@med.unc.edu

The P2Y1 receptor (P2Y1-R) is expressed broadly in mammalian tissues, for example on the surface of platelets where it is essential for ADP-promoted aggregation. Our laboratories have focused on the development of P2Y1- R-selective antagonists, agonists, and radioligands as reagents for delineating the physiological and molecular properties of this important cell signaling protein. Synthesis of a series of selective, high affinity, non-nucleotide competitive antagonists for the P2Y1-R led to development of [3H]MRS2279 as a radioligand for reliable screening of agonists and antagonists of the P2Y1 receptor. Moreover, the human P2Y1-R was purified to homogeneity taking advantage of this radioligand and high level expression of recombinant receptor from a baculovirus in insect cells. We recently developed methodology for radiolabelling with 32P the highest affinity (Kd ∼ 0.5 nM) P2Y1-R antagonist, 2-iodo-N6-methyl-(N)-methanocarba-2′-deoxyadenosine-3′,5′-bisphosphate (MRS2500). This high specific radioactivity, high affinity radioligand has proved very useful for quantification of natively expressed P2Y1-R across mammalian tissues. Moreover, we have applied this radioligand to quantify native P2Y1-R during agonist-induced trafficking in human platelets and in cell lines. Structure activity studies of P2Y1-R antagonists also were instrumental in development of (N)-methanocarba-2MeSADP (MRS2365), which is a high affinity full agonist at the ADP-activated P2Y1-R but does not bind to the ADP-activated P2Y12-R or P2Y13-R. This molecule provides a reagent for specific activation of the P2Y1 receptor in tissues, e.g. platelets, where multiple ADP-activated receptors exist.

Release of nucleotides and UDP-sugars from Airway Epithelia

Eduardo R. Lazarowski

Cystic Fibrosis Center. University of North Carolina School of Medicine Eduardo_Lazarowski@med.unc.edu

Extracellular nucleotides modulate multiple components of the innate lung defense by activating epithelial airway surface purinergic receptors. ATP, UTP, UDP, and adenosine control the production of airway surface liquid (ASL) by regulating ion transport. ASL nucleotides also promote cilia beating and mucin secretion, and thereby activate the mucociliary clearance (MCC) process that removes noxious materials from the airways. In addition, a recent study illustrated that the P2Y14 receptor (the cognate receptor for UDP-glucose, UDP-Glc) is expressed in alveolar epithelial type II and other lung epithelial cells and promotes the secretion of the potent neutrophil chemoattractant IL-8 (1). Thus, UDP-Glc may be an inflammatory mediator in the airways. Noteworthy, P2Y14 receptor transcripts are most abundantly expressed in circulating neutrophils, relative to other peripheral tissues (2). Quantitative PCR analysis on human neutrophils illustrated the presence of P2Y14 receptor transcripts in copy number comparable to that of the abundantly expressed chemotactic peptide (fMLP) receptor FPR-1.

Despite the physiological and pathophysiological relevance of the responses triggered by extracellular nucleotides and UDP-sugars in the airways, the mechanisms and pathways for nucleotide release from airway epithelia are unknown. UDP-sugars are donor substrates for glycosylation reactions in the lumen of the secretory pathway. UDP is a byproduct of this process, which in turn is hydrolyzed to UMP by a Golgi resident UDPase. UDP-sugar/UMP antiporters translocate luminal UMP to the cytosol and cytosolic UDP-sugars to the lumen of the Golgi. We have discovered that in addition to ATP, airway epithelial cells release UDP-Glc constitutively and that release of UDP-Glc from resting epithelial cells is markedly affected by genetic manipulation of Golgi UDPsugar translocators. These observations suggest that UDP-Glc is released to the extracellular milieu from the secretory pathway, i.e., following translocation to the Golgi via UDP-sugar/UMP antiporters.

Impaired MCC, mucus hypersecretion, and neutrophil inflammation are characteristics of chronically diseased airways, e.g., cystic fibrosis (CF), chronic obstructive lung disease (COPD), and asthma. UDP-Glc levels in sputum samples from CF patients are markedly elevated, reaching concentrations (∼500 nM) capable of promoting robust activation of P2Y14 receptors. Unlike ATP, UDP-Glc added either onto the mucosal surface of airway epithelial cells or to CF sputum was poorly hydrolyzed. In addition, enhanced ATP and UDP-Glc release occurs during Ca2+-promoted vesicle exocytosis and mucin granule secretion from airway epithelial goblet cells (3). Therefore, nucleotide release from goblet cells may be mechanistically associated with mucin secretion, e.g., as co-cargo molecules within mucin granules.

Together, our observations suggest that (i) airway epithelial cell nucleotide release has an exocytotic component, and (ii) sustained UDP-Glc accumulation in ASL during conditions associated with mucin hypersecretion e.g., CF, COPD, and asthma, provides long lasting pro-inflammatory signaling in the airways.

Role of diadenosine polyphosphates in corneal wound healing

Jesus Pintor1, Aranzazu Mediero1 and Assumpta Peral2

1Dto. Bioquímica y Biología Molecular IV, 2Dep. Óptica II, E. U. de Óptica Universidad Complutense de Madrid, Spain. jpintor@vet.ucm.es

As the epithelium is the most external layer of the cornea, it is often damaged by several factors causing corneal wounds. These wounds are repaired in a three-step process called corneal wound healing. Diadenosine polyphosphates, ApnA, are present in rabbit and human tears. This fact invites to think that these dinucleotides may participate in ocular surface processes such as corneal wound healing. Therefore, we have investigated the possible role of diadenosine polyphosphates on corneal wound healing by studying the changes in the rate of corneal epithelial cell migration after dinucleotide applications.

Primary corneal epithelial cell cultures were obtained from New Zealand white rabbits. Immunocytochemical experiments were carried out by fixing the cells with 4% PFA and incubated with citokeratine 3 primary antibody which was subsequently incubated with a secondary Ig-G mouse labelled with FITC an cells were observed under confocal microscopy. Migration studies were carried out by taking confluent monolayers which were wounded with a pipette tip and challenged with different di- and mononucleotides (Ap3A, Ap4A, Ap5A, Up4U, ATP, UTP, ADP and UDP). For concentration-response analysis compounds were tested in doses ranging from 10−8 to 10−3 M. When the P2 antagonists, PPADS, suramin and reactive blue 2, were used, they were assayed at 100 µM, and from 10−7 to 10−3 M in concentration-response studies. In order to study the intracellular pathways involved in cell migration, several MAPK and citoskeleton inhibitors (U0126 100 µM, Y27632 100 2µ, AG1478 100 2µ, PAO (phenylarsine oxide) 5 µM, (−)-Blebblistatin 10 µM and ML7 25 µM) were assayed in the presence or absence of Ap4A and Ap3A both 100 µM. Stability of the dinucleotides was assayed by HPLC using an isocratic method.

Cells under study were identified as corneal epithelial cells due to this positive labelling to cytokeratine-3 in the immunocytochemical analysis. Cell migration experiments showed that Ap4A, UTP and ATP accelerate the rate of healing, while Ap3A, Ap5A and UDP delay it. ADP and Up4U did not modify the rate of migration. For further experiments we took Ap4A (which accelerates the rate of wound) and Ap3A which delays it. P2Y antagonists presented small differences between Ap4A and Ap3A. The assays with MAPK and citoskeleton inhibitors, revealed that both, MAPK pathway and cdc-42/RAC/RhoA/ROCK pathways are involved in the epithelial migration. In this sense, Ap4A activating cdc-42 cascade increase the rate of corneal epithelial cell migration, while the delay caused by Ap3A is mainly due to the MAPK pathway Finally, concerning the possible degradation of the dinucleotides it was almost impossible to detect any product as a consequence of their cleavage. Degradation of 1–3% of the dinucleotides in 2 min perfusion indicates that the active molecules are the diadenosine polyphosphates and not the generated mononucleotides. In summary, analysing the pharmacological profile of all the compounds tested we can conclude that diadenosine polyphosphates activate two main P2Y receptors: a P2Y2 receptor accelerating the rate of healing and a P2Y6 receptor which delays this process.

Medicinal Chemistry

Adenine nucleosides as reversible P2Y12 antagonists

James G. Douglass, Adam Samuelson, J. Bryan deCamp, Emilee Glaub, Dima Smirnov, Sanjoy Mahanty, Anna Morgan, Chris Crean, Jose L. Boyer, Stephanie Anderson, and Paul S. Watson

Inspire Pharmaceuticals Inc., 4222 Emperor Boulevard, Suite 200, Durham, North Carolina, USA 27703-8466 pwatson@inspirepharm.com

ADP serves as an endogenous agonist for both P2Y1 and P2Y12 on platelets, and antagonism of its action at either receptor has been shown to inhibit platelet aggregation. Inhibition of platelet aggregation using irreversible P2Y12 antagonists has proven to be a useful therapy to treat Acute Coronary Syndrome (ACS). Previously we had identified several lipophilic modifications to adenine nucleotides that enable them to reversibly block aggregation mediated by P2Y12. This presentation will briefly discuss human clinical trial results of one of these reversible P2Y12 adenine nucleotide antagonists, INS50589, and the discovery of reversible P2Y12 adenine nucleoside antagonists stemming from observations gained from our original nucleotide research.

Development of pharmacological tools for studying P2Y receptors

C.E. Müller

Pharmaceutical Sciences Bonn (PSB), Pharmaceutical Chemistry, Pharmaceutical Institute Poppelsdorf, University of Bonn, Kreuzbergweg 26, 53115 Bonn, Germany christa.mueller@uni-bonn.de

P2Y receptors are G protein-coupled receptors (GPCRs) activated by nucleotides, such as ATP, ADP, UTP, UDP, or UDPglucose, depending on the receptor subtype. Currently, 8 subtypes are known (physiological agonist in brackets): P2Y1 (ADP), P2Y2 (UTP, ATP), P2Y4 (UTP), P2Y6 (UDP), P2Y11 (ATP), P2Y12 (ADP), P2Y13 (ADP), and P2Y14 (UDPglucose) [1].

Our project focussed on the synthesis of base-modified uracil nucleotides and metabolically stable nucleotide mimetics as ligands for the uracil nucleotide-sensitive P2Y receptor subtypes. A series of nucleotides and analogs was synthesized, investigated at recombinant human P2Y2, P2Y4, and P2Y6 receptors, and structure-activity relationships were analyzed. Alkyl substituents in the 6-position of UTP and UDP were not tolerated by the three receptor subtypes. Large, aromatic substituents at N3 of UDP were well tolerated by the P2Y6 receptor leading to potent, P2Y6-selective agonists. Modification of the triphosphate chain by introduction of β,γ-dichloromethylene bridge in UTP derivatives, resulting in compounds with enhanced metabolic stability, was well tolerated by all three receptor subtypes.

Radioligands are valuable tools for studying GPCRs, however, selective radioligands have only been available for the P2Y1 receptor subtype so far. We have recently developed the first selective radioligand for the P2Y12 receptor subtype [2], an ADP receptor predominantly expressed on platelets, and in lower density in the brain. The P2Y12 antagonist radioligand [3H]PSB-0413 (2-propylthioadenosine-5′-adenylic acid (1,1-dichloro-1-phosphonomethyl-1-phosphonyl) anhydride, AR-C67085MX) has been successfully used for characterizing P2Y12 receptors on the protein level in various tissues and cells. The radioligand allowed to set up a fast and efficient screening assay for identifying novel lead structures for the development of P2Y12 antagonists, which are potent antithrombotic agents.

Extracellular nucleotides are quickly hydrolyzed by ectonucleotidases to limit their action. Inhibitors of such enzymes would act as indirect P2Y receptor agonists by prolonging the effects of endogenously released nucleotides. We have developed a fast and convenient nanoscale capillary electrophoresis assay to screen and characterize inhibitors of ectonucleotidases [3]. The enzymatic reaction is performed within the capillary followed by electrophoretic separation and quantification (by UV) of the reaction products. A novel class of ectonucleotidase inhibitors has been discovered by this approach.

Molecular Recognition in P2Y Nucleotide Receptors

Kenneth A. Jacobson

Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA kajacobs@helix.nih.gov

Agonists selective for P2Y1, P2Y2, and P2Y6 receptors and nucleotide antagonists selective for P2Y1 and P2Y12 receptors have been identified. Selective non-nucleotide antagonists have been reported for P2Y1, P2Y2, P2Y6, P2Y11, P2Y12, and P2Y13 receptors. Three-dimensional structural models of P2Y receptors, deduced mainly from rhodopsin-based homology modeling and mutagenesis studies, have aided in the development of selective ligands [1]. We have synthesized, as ligands for P2Y receptors, nucleotide analogues in which the ribose moiety is substituted by a variety of novel ring systems, including conformationally-locked moieties. The focus on conformational factors of the ribose-like moiety allows the inclusion of general modifications that lead to enhanced potency and selectivity. In solution, the ribose ring of an unbound nucleotide may exist in a dynamic equilibrium between (N) (North; 2′-exo/3′-endo) and (S) (South; 2′-endo/3′-exo) conformations, and X-ray crystallographic structures of diverse nucleotide complexes through nature indicate a clustering around these conformations. It is possible to stabilize each of these conformational clusters by chemical bridging within a ring. Replacement of the ribose moiety of ATP with a bicyclo[3.1.0]hexane ring (methanocarba, mc) ring system, i.e. fused cyclopropane and cyclopentane rings, locks the analogue in either a (N) or (S) conformation, depending on the position of the CH2-bridge. These two isomeric variants produce agonist analogues having widely differing activities at P2 receptors (collaboration w/T.K. Harden). At P2Y1,2,4,11 receptors, there is a preference for the (N) conformation as indicated with (N)-methanocarba analogues, such as the potent, competitive P2Y1 antagonist MRS2500. MRS2500 was sufficiently stable in vivo to inhibit ADP-induced platelet aggregation in mice [2]. [33P]MRS2500 was synthesized enzymatically and studied as a radioligand having a Kd value of 0.3 nM at the rat brain P2Y1 receptor [3]. MRS2365, an (N)-methanocarba analogue of 2-MeSADP, displayed potency (EC50) of 0.4 nM at the P2Y1 receptor with >10,000-fold selectivity in comparison to P2Y12 and P2Y13 receptors [3]. Basic structure activity relationships were explored at P2Y2 and P2Y4 receptors. However, there is a fundamental conformational difference between the binding sites of P2Y6 and various other P2Y receptors. A uridine 5′-diphosphate analogue locked in the (N) envelope conformation was inactive [4]. Based on a prediction from docking of nucleotide derivatives to a P2Y6 receptor model, (S)-mc-2′-deoxy-UDP was synthesized and found to be more potent than the corresponding riboside, dUDP, indicating a preference for the (S) conformation. At the recombinant human P2Y13 nucleotide receptor, MRS2211 (6-[(2-chloro-5-nitrophenyl)-azo]-pyridoxal-5′-phosphate) an analogue of the known P2 receptor antagonist PPADS was found to competitively and selectively inhibit functional activity with a pIC50 value of 5.97, being 45-fold more potent than PPADS [5].

New Nucleotides and Nucleotides Analogs and their Activity on P2 Receptors

Gloria Cristalli and Rosaria Volpini

Dipartimento di Scienze Chimiche, Università di Camerino, via S. Agostino, 1, 62032 Camerino, Italy Presenting author: gloria.cristalli@unicam.it

P2 receptors, activated by nucleotides like ATP, ADP, UTP, and UDP, belong to two families: the ionotropic P2X receptors (P2X1–7), which represent a class of ligand-gated ion channels and the metabotropic P2Y receptors (P2Y1,2,4,6,11–14), which are ubiquitously expressed in the human body. All these receptors play important pathophysiological roles, hence, the availability of new ligands can provide novel drugs for the treatment of a number of currently-incurable diseases including tumors and neurodegenerative disorders. However, a limited number of reports described the identification of potent and selective P2X/P2Y ligands. Hence, the present work will be focused on the design and characterization of novel P2 receptor ligands obtained through the phosphorylation of substituted purine and pyrimidine nucleosides [1]. Furthermore, the design of “mini nucleotides” based on the adenine skeleton, and their preparation using solid phase organic synthesis (SPOS) will be presented. Open image in new window

Studies on different pharmacological models such as: a) effects on the human platelet aggregation, b) neuronal differentiation and commitment to death in the human neuroblastoma SH-SY5Y cells, and c) activity on HEK 293 cells transfected by ionotropic P2X3 receptors, will be discussed. Preliminary results demonstrated that the above mentioned nucleotides interact with the P2 receptors; in particular, some 2-alkynyladenosine nucleotides behave as agonists or antagonists at platelet P2Y1 and P2Y12 subtypes, depending on the nature of the substituent in 2-position [2], while the 5-iodouridine triphosphate interacts with the metabotropic P2Y4 receptor subtype leading to neuroblastoma SH-SY5Y cell death. Furthermore, some mininucleotides demonstrated agonist activity on P2X3 receptors.

Novel orthosteric and allosteric ligands for adenosine receptors

Ad P. IJzerman

Leiden/Amsterdam Center for Drug Research, PO Box 9502, 2300RA Leiden, The Netherlands (email: ijzerman@lacdr.leidenuniv.nl)

Adenosine receptor antagonists usually possess a bi- or tricyclic heteroaromatic structure at their core with varying substitution patterns to achieve selectivity and/or greater affinity. Taking into account molecular modelling results from a series of potent A1 adenosine receptor antagonists, we derived a pharmacophore suggesting that a monocyclic core can be equally effective. As a further design criterion we imposed a restriction on the polar surface area (PSA) of the molecules that would allow them to penetrate into the CNS. In consequence, we have synthesised two novel series of pyrimidines and a related series of purines with unanticipated substitution pattern, possessing good antagonistic potency at the A1 adenosine receptor and desirable PSA values. In particular, pyrimidine LUF5735 and purine LUF5962 display excellent, even subnanomolar affinity and selectivity at the human A1 receptor. Open image in new window

As a next challenge, a series of 1H-imidazo-[4,5-c]quinolin-4-amine derivatives was synthesized as allosteric modulators of the human A3 adenosine receptor. Structural modifications were made at the 4-amino and 2-positions. The compounds were tested in both binding and functional assays, and many were found to be allosteric enhancers of the action of A3AR agonists by several different criteria. First, a potentiation of the maximum efficacy of the agonist Cl-IB-MECA was observed for numerous derivatives, with the greatest increase of 45–50% observed for LUF6000. Also, a number of these compounds, including LUF6000, decreased the rate of dissociation of the agonist [125I]I-AB-MECA from the A3 receptor. It was found that the capability of these compounds to increase agonist efficacy correlated with their ability to decrease the dissociation rate, but not their inhibition of equilibrium binding.

In conclusion we have prepared and evaluated novel series of compounds that display remarkable characteristics at either human adenosine A1 or A3 receptors.

Recent developments in the field of A2B and A3 adenosine receptors antagonists

Pier Giovanni Baraldi1, Delia Preti1, Mojgan Aghazadeh Tabrizi1, Francesca Fruttarolo1, Romeo Romagnoli1, Naser Abdel Zaid3, Allan R. Moorman4, Stefania Merighi2, Katia Varani2, and Pier Andrea Borea2

1Dip. Scienze Farmaceutiche, 2Dip. di Medicina Clinica e Sperimentale-Sezione di Farmacologia, Università di Ferrara, 3College of Pharmacy, An-Najah National University, Nablus, 4King Pharmaceutical Research and Development, Inc., 4000 CentreGreen Way, Suite 300, Cary, North Carolina 27513

Adenosine, an endogenous modulator of a wide range of biological functions in the nervous, cardiovascular, renal, and immune systems, interacts with at least four cell surface receptor subtypes classified as A1, A2A, A2B and A3 [1]. Clarification of the role of adenosine and its receptors in cancer development may hold great promise for the chemotherapeutic treatment of patients affected by malignancies [2].

Different classes of compounds with non-xanthine structures have been reported to be A3 adenosine receptor antagonists [3]. The pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine nucleus has been largely investigated by our group. Our interests were focused on the effects of substitution of the phenyl ring of the arylcarbamoyl moiety at N5 position and of substituents at C9 and/or at N7-N8 pyrazole nitrogens. These studies allowed us to obtain a large variety of compounds which showed affinities in the nanomolar range to human A2A or A3 adenosine receptors with high degree of selectivity [3].

Compounds presenting an additional fused ring on the xanthine nucleus have been reported to exhibit antagonistic activity with various levels of affinity and selectivity towards the four adenosine receptors subtypes [4]. We evaluated the effect of the introduction of a benzyl and a propyl at the 1 and 3 positions, respectively, in a new series of 7-aryl/alkyl-1H,6H-pyrrolo[2,1-f]purine-2,4-diones and 7-aryl/alkyl-1H,8H-imidazo[2,1-f]purine-2,4-diones [5], among which, very potent and selective A3 receptors antagonists have been identified. In particular 1-benzyl-7-methyl-3-propyl-1H,8H-imidazo[2,1-f]purine-2,4-dione, shows a subnanomolar affinity with a noteworthy selectivity versus the other adenosine receptors subtypes (Ki (hA3) = 0.8 nM, Ki (hA1/hA3) = 3163, Ki (hA2A/hA3) > 6250, IC50 (hA2B)/Ki (hA3) = 2570).

Colotta et al. directed much effort toward the study of adenosine receptor antagonists investigating the 2-arylpyrazolo[3,4-c]-quinoline nucleus [6]. In light of the reported activity profile, we decided to synthesize the structural isomers, 2-arylpyrazolo[4,3-c]quinolines [7], some of which showed high A3 receptor affinity and complete selectivity (2-p-tolyl-2,5-dihydro-pyrazolo[4,3-c]quinolin-4-one; KihA1, KihA2A, EC50hA2B>1000 nM, KihA3= 9 nM).

In the search for improved selective A2B antagonists for the treatment of asthma [8], we synthesized a variety of new 1,3-dipropyl-8-heterocyclic-substituted xanthines [9]. We introduced several heterocycles, such as pyrazole, isoxazole, pyridine and pyridazine at the 8-position of the xanthine nucleus. We have also investigated different spacers (substituted acetamide, oxyacetamide and urea moieties) on the heterocycle introduced. Some of the synthesized C8-substituted xanthines showed high affinity at A2B receptor subtype and very good selectivity (Nbenzo[1,3]dioxol-5-yl-2-[5-(2,6-dioxo-1,3-dipropyl-2,3,6,7-tetrahydro-1H-purin-8-yl)-1-methyl-1H-pyrazol-3-yloxy]-acetamide; hA2B = 5.5 nM, hA1, hA2A, hA3 > 1000).

Synthesis and molecular modeling studies of new pyrazolo[3,4-b]pyridines, potent and selective inhibitors of A1 adenosine receptors

Maurizio Botta1, Fabrizio Manetti1, Cristina Tintori1, Adriano Martinelli2, Tiziano Tuccinardi2, Letizia Trincavelli2, Claudia Martini2, and Silvia Schenone3

1Dipartimento Farmaco Chimico Tecnologico, Università degli studi di Siena, Via Aldo Moro, 3, 53100 Siena, Italia;

2Dipartimento di Scienze Farmaceutiche, Università degli studi di Pisa, Via Bonanno Pisano, 6, 56126 Pisa, Italia;

3Dipartimento di Scienze Farmaceutiche, Università degli studi di Genova, Viale Benedetto 15, 16132 Genova, Italia botta@unisi.it

Adenosine is an endogenous neuromodulator, which mediates its biological effects by interacting with four receptor subtypes named A1, A2A, A2B, and A3. The physiological significance and functions of adenosine have been extensively studied and in the last twenty years a large number of adenosine ligands, agonists and antagonists, have been developed. Particularly, many efforts have been invested in the synthesis of A1ARs antagonists, that can stimulate cerebral activity by blocking the adenosine central inhibitory activity. In this context, a number of new 4-amino-1-(2-chloro-2-phenylethyl)-1H-pyrazolo[3,4-b]pyridines 5-carboxylic acid esters have been synthesized by our research group and their binding affinity at the A1, A2A and A3 adenosine receptors (AR) have been evaluated. The new compounds were characterized by a high affinity and selectivity toward the bovine A1 receptor subtype (bA1AR). In addition, to rationalize the relationships between structure and affinity of such compounds together with A1AR antagonists taken from the literature, molecular modeling studies were carried out. As a first approach, a pseudoreceptor model [1] was built to derive a hypothesis of the interaction pathway between the set of A1AR antagonists and a model of the putative A1 receptor. Nevertheless, biological evaluation of several compounds suggested by the model did not give the expected results, demonstrating that the model itself was scarcely able to predict the activity of inhibitors subjected to modifications at the position 5. As a consequence, a new and more accurate model was required to better predict the activity of A1AR inhibitors. To reach this goal, a highly predictive 3D QSAR model for antagonists of A1AR was generated [2]. Such model demonstrated to have the knowledge for estimating the affinity of a very large number of inhibitors belonging to different structural classes. The generated 3D QSAR has been also used to refine a model of the A1AR receptor built by homology and then molecular modeling study of the interaction of the compounds under investigation with the bovine A1 adenosine receptor was carried out. On the other hand, the ADME (absorption, distribution, metabolism, excretion) properties of the synthesized pyrazolo-[3,4-b]pyridines were evaluated by means of the Volsurf software with the aim to have a prediction of the aqueous solubility and cell permeability of such molecules. All computational studies together with the synthesis of new pyrazolo-[3,4-b]pyridines will be reported during the oral presentation and the biological evaluation of the inhibitors on both the bovine and human A1 receptors (bA1AR and hA1AR respectively) will be also discussed.

The Discovery of Selective, High Affinity A2B Adenosine Receptor Antagonists

Jeff Zablocki1, Rao Kalla1, Elfatih Elzein1, Thao Perry1, Xiaofen Li1, Venkata Palle1, Vaibhav Varkhedkar1, Art Gimbel2, David Lustig3, Kwan Leung3, Tim Glennon4, Hugh Genin4, and Dewan Zeng2

1Department of Bioorganic Chemistry, 2Department of Drug Research and Pharmacological Sciences, 3Department of Pre-Clinical Development CV Therapeutics Inc., 3172 Porter Drive, Palo Alto, CA 94304, USA Accelrys, Inc.

The P1 family of adenosine receptors consists of two Gi couple receptors, A1 and A3, and two Gs coupled receptors, A2A and A2B. The A2B adenosine receptor (AdoR) is found in multiple tissues including the lung where it is found in bronchial smooth muscle cells [1] and mast cells [2]. In vitro experiments, demonstrate the release of inflammatory cytokines from both bronchial smooth muscle cells and mast cells, effects that can be blocked by an A2B AdoR antagonist [1,2]. Thus, we hypothesize that the A2B AdoR may play a role in asthma. We will disclose an overview of our A2B AdoR antagonist structure activity relationships starting with MRS-1754 [3] as a lead molecule to the discovery of 8-(1H-pyrazol-4-yl)xanthine 1, a compound with a 1 nM Ki for the A2B AdoR and excellent selectivity (Figure 1). The complete SAR leading up to the discovery of 1 involved a systematic optimization of the heterocycle at the 8 position, followed by varying the chain length of the N-1 aralkyl (phenpropyl, phenethyl, and benzyl), optimizing substituents on the phenyl ring for affinity and selectivity for the A2B AdoR, and finally varying the N-1 and N-3 alkyl substituents. Unfortunately, 1 did not possess high oral bioavailability (rats), and this was attributed to low water solubility. Therefore, an effort to add charge to the molecule to enhance water solubility was explored. Initially, adding a carboxylate to the phenyl ring or benzylic methylene was explored, and the latter approach afforded a compound 2 of modest A2B AdoR affinity and selectivity (Figure 1). However, upon oral administration of 2 to rats it was rapidly converted to the unsubstituted pyrazole, a non-selective A2B antagonist. Fortunately, replacement of the phenyl group with pyridyl as in 4 resulted in high affinity for the A2B AdoR and modest selectivity. Furthermore, 4 demonstrated improved oral bioavailability following dosing in rats. A homology model of the A2B AdoR receptor was built using MODELER based on the X-Ray of bovine rhodopsin. Various docking techniques (Ligand-Fit and Affinity), were used to develop a hypothetical binding pose for 1 and 4 consistent with the above SAR and published site-directed mutagenesis data, and the outcome of these efforts will be presented. Open image in new window

Cardiovascular System

Adenosine and Vasomotor Regulation in the Coronary Microcirculation

Lih Kuo1 and Travis W. Hein2

1Department of Systems Biology and Translational Medicine, and 2Department of Surgery, Scott & White Memorial Hospital, College of Medicine, Texas A&M University System Health Science Center, Temple, TX 76504, USA. LKUO@tamu.edu

The nucleoside adenosine has been suggested to play a major role in the regulation of coronary blood flow during metabolic stress (i.e., hypoxia and ischemia) by the preferential dilation of small coronary arterioles primarily through the activation of adenosine A2A receptors. However, the signaling pathways leading to this dilation are incompletely understood. Since tissue acidosis and alteration in local hemodynamics are associated with metabolic stress, it is important to examine whether the adenosine-induced vasodilatory response can be influenced by these factors. Using an isolated vessel approach and videomicroscopic techniques, these issues were addressed recently in our laboratory. It was found that disruption of endothelium, blockade of endothelial ATP-sensitive potassium (KATP) channels by glibenclamide, inhibition of nitric oxide (NO) synthase by L-NAME and of soluble guanylyl cyclase by ODQ produced identical attenuation of vasodilation to adenosine. Combined administration of these inhibitors did not further attenuate the vasodilatory response. Pertussis toxin (PTX), but not cholera toxin, significantly inhibited vasodilation to adenosine, and this inhibitory effect was only evident in vessels with an intact endothelium. Both glibenclamide and a high concentration of extraluminal KCl abolished vasodilation of denuded vessels to adenosine; however, inhibition of calcium-activated potassium channels had no effect on this dilation. Rp-8-Br-cAMPS, a cAMP antagonist, inhibited vasodilation to cAMP analog 8-Br-cAMP but failed to block adenosine-induced dilation. These results suggest that adenosine activates both endothelial and smooth muscle pathways to exert its vasomotor activity. On one hand, adenosine opens endothelial KATP channels through activation of PTX-sensitive G-proteins. This signaling leads to the influx of Ca2+ for NO synthesis, which subsequently increases smooth muscle cGMP for vasodilation. On the other hand, adenosine activates smooth muscle KATP channels and leads to vasodilation through hyperpolarization. The latter pathway is independent of G-proteins and cAMP/cGMP signaling. In the presence of a threshold concentration of adenosine (0.1 nM), the vasodilation in response to increased lumenal flow was significantly enhanced. Inhibiting endothelial KATP channels (by lumenal glibenclamide) or depolarizing endothelial cells (by lumenal KCl, 40 mM) abolished the effect of adenosine. Reducing intraluminal pressure from 80 to 40 cmH2O significantly enhanced vasodilations to adenosine and pinacidil (KATP channel activator). These enhanced responses were not affected by endothelial removal but were abolished by a subthreshold concentration of glibenclamide. In contrast to adenosine, sodium nitroprusside (SNP)-induced dilation was not affected by pressure changes. In the acidic solution (pH = 7.3), the dilations to both adenosine and pinacidil, but not to SNP and KCl (10 to 20 mM), were enhanced in a glibenclamide-sensitive manner. In summary, these findings suggest that adenosine elicits coronary arteriolar dilation by activating KATP channels for membrane hyperpolarization in the endothelial (i.e., NO release) and smooth muscle cells. Adenosine, by activating endothelial KATP channels, can potentiate coronary arteriolar dilation in response to increased flow. In addition, activation of smooth muscle KATP channels, by a mild acidosis or lowering intraluminalpressure,increasescoronaryarteriolar dilation to adenosine. It appears that adenosine can interact with local hemodynamics and chemical factors for integrative vasomotor regulation by selectively modulating resting KATP channel activity in either endothelial or smooth muscle cells.

An Emerging Role for Adenosine in Angiogenesis

Thomas H. Adair

Department of Physiology & Biophysics, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216, USA tadair@physiology.umsmed.edu

It is well-established that metabolic factors have a critical role in the regulation of angiogenesis [1]. An increase in metabolic activity leads to a decrease in tissue oxygenation causing tissues to become hypoxic. The hypoxia initiates a variety of signals that stimulate angiogenesis, and the increase in vascularity that follows promotes oxygen delivery to the tissues. When the tissues receive adequate amounts of oxygen even during peak levels of activity, the intermediate effectors return to normal levels, and angiogenesis ceases.

An emerging concept is that adenosine released from hypoxic tissues has an important role in driving the angiogenesis [2, 3]. The following feedback control hypothesis is proposed: AMP is dephosphorylated by ecto-5′-nucleotidase, producing adenosine under hypoxic conditions in the extracellular space adjacent to a parenchymal cell (eg, cardiomyocyte, skeletal muscle fiber, hepatocyte, etc.). Extracellular adenosine activates A2 receptors, which stimulates release of vascular endothelial growth factor (VEGF) from the parenchymal cells. VEGF binds to its receptor (VEGFR-2) on endothelial cells, stimulating their proliferation and migration. Adenosine can also stimulate endothelial cell proliferation independently of VEGF, which probably involves modulation of other proangiogenic and antiangiogenic growth factors, and perhaps an intracellular mechanism. In addition, hemodynamic factors associated with adenosine-induced vasodilation may have a role in development and remodelling of the vasculature. Once a new capillary network has been established, and the diffusion/perfusion capabilities of the vasculature are sufficient to supply the parenchymal cells with adequate amounts of oxygen, adenosine and VEGF as well as other proangiogenic and antiangiogenic growth factors return to near normal levels, thus closing the negative feedback loop. The available data indicate that adenosine might be an essential mediator for up to 50–70% of the hypoxia-induced angiogenesis in some situations; however, additional studies in intact animals will be required to understand fully the quantitative importance of adenosine.

Adenosine stimulation of angiogenesis is a natural phenomenon that occurs in the microenvironment of tissues in accordance with local metabolic needs. Amplification of this natural process through modulation of the enzymes of adenosine metabolism, or genetic or pharmacological induction of A2A or A2B receptors could, theoretically, lead to stimulation of angiogenesis in those areas of a tissue where hypoxia is most severe and adenosine levels are highest. This type of physiological therapy can be viewed as a departure from classical attempts to stimulate angiogenesis in a global manner, in which many or most cells of an organ or tissue are made to produce greater amounts of VEGF. In contrast to adenosine-based therapies, global induction of VEGF may not be an effective means to build capillary networks in the microenvironment of a tissue where hypoxia is most severe. (Supported by National Heart, Lung, and Blood Institute, HL51971.)

Control of extracellular adenosine concentration by ecto-5′-nucleotidase

A. Deussen, A. Pexa, J. Weichsel, S. Sternberger

Department for Physiology, Medical Faculty Carl Gustav Carus, TU Dresden Adenosine is produced inside the cell by action of 5′-nucleotidase(s) and S-adenosylhomocysteine hydrolase. In addition, there is compelling evidence for extracellular adenosine production by ecto-5′-nucleotidase. A previous analysis of complex data sets of cardiac adenosine production has provided strong evidence that the prevailing path of adenosine production in the heart is intracellular (Deussen et al., Circulation 1999; 99: 2041–47). However, it was also found that intracellular removal of adenosine via adenosine kinase and adenosine deaminase was very rapid, and for physiological baseline conditions a transmembraneous adenosine concentration gradient was predicted, which was directed from extra- to intracellular. This scenario opens the possibility that extracellular adenosine production by ecto-5′-nucleotidase (CD73) is an important source of adenosine, which may significantly modulate the extracellular adenosine concentration.

This hypothesis was tested in isolated endothelial cells and isolated perfused mouse hearts with respect to an activation of CD73, pharmacological block of the enzyme and a genetic knock out approach. Baseline activity of CD73 was assessed by measurement of the conversion of 1,N6-etheno-AMP to 1,N6-etheno-adenosine in intact cells or hearts (HPLC assay with fluorescence detection). While conversion of the etheno-labeled AMP provided a read out of CD73 activity of intact cells or tissues, additional measurement of native adenosine release (HPLC-technique using UV-detection) gave information on the production of native (endogenous) adenosine. Application of lyso-phospholipids (lysophosphatidyl-choline,10 µM or platelet activation factor, 5 µM) enhanced catalytic activity of CD73 of micro- and macrovascular endothelial cells. Moreover, in presence of blockers of membrane transport inhibitors (nitrobenzylthioinosine 1 µM, dipyridamole 1 µM) the native extracellular adenosine concentration was doubled within 2 min after application of lyso-phospholipids. In contrast, block of CD73 by α,β-methylene-ADP (50 µM) significantly decreased the extracellular adenosine concentration and completely blocked the effect of lyso-phospholipids on CD73 activity. In isolated hearts from CD73 −/− mice the adenosine concentration in the venous effluent perfusate was only 1/3 of the concentration measured in hearts of WT mice under identical conditions (pO2 > 600 mm Hg, perfusion pressure 85 mm Hg).

We conclude that ecto-5′-nucleotidase (CD73) represents a key enzyme for modulation of extracellular adenosine concentration in the cardiovascular system. This extracellular adenosine production may be of particular importance for conditions when total adenosine production is low. Lyso-phospholipids are of potential physiological or pathophysiological importance, because they can almost instantly enhance the activity of ecto-5′-nucleotidase resulting in increased extracellular adenosine concentrations.

The forearm model as a tool to study the clinical pharmacology of adenosine in humans

Gerard A. Rongen, M.D., Ph.D.

Departments of Pharmacology-Toxicology and Internal Medicine, Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands G.Rongen@pharmtox.umcn.nl

Preclinical evidence identifies adenosine as a ‘retaliatory metabolite’, protecting cells against the deleterious consequences of ischemia and limited nutrient supply. Translation of this knowledge to humans in-vivo is needed because of inter-species differences in the pharmacology of adenosine. Various challenges are encountered in this translational effort including (1) extremely short plasma half-life of adenosine due to rapid cellular uptake and metabolism, preventing circulating adenosine from reaching its target cells, (2) adenosine-induced autonomic reflexes that mask the local actions of adenosine at autonomic nerve endings, vascular smooth muscle cells and cardiomyocytes [1] and (3) the lack of tools to study ischemic injury in an experimental setting in humans. Various strate-gies have successfully been applied to meet these challenges. The forearm model has proven its value in this regard. In this model, the brachial artery is cannulated for drug infusion and blood sampling. Intra-arterial delivery of drugs provides a concentration gradient between infused forearm and systemic circulation, allowing exploration of local effects of adenosine with minimal interference by autonomic reflexes or systemic actions [1, 2]. Using forearm strain gauge plethys-mography, the influence of adenosine on forearm blood flow can be studied. In this model, we have successfully demonstrated that the vasodilator property of adenosine is mediated by adenosine re-ceptors, and involves nitric oxide [3]. Using dipyridamole to inhibit cellular uptake of adenosine, we provided support in humans in-vivo for the hypothesis that KATP-channels are involved in adeno-sine-induced vasodilation [4] and that extracellular adenosine levels are reduced in patients with hyperhomocysteinemia but increased during oral treatment with methotrexate [5, 6]. When the model is further refined by cannulation of an antecubital vein, effects on local and systemic release of endogenous substances such as norepinephrine can be explored. In this way, we have demonstrated that adenosine reduces norepinephrine release from sympathetic nerve endings [2, 7]. Recently, we have developed forearm annexin A5 scintigraphy to quantify forearm ischemiareperfusion (I/R) injury. With this technique, we detected protection against I/R injury by ischemic preconditioning, local infusion of adenosine and oral treatment with dipyridamole [8, 9]. Future research will charac-terize environmental and genetic factors that determine protection against I/R injury by adenosine.

Molecular mechanisms of coronary flow regulation by adenosine

S. Jamal Mustafa

Department of Physiology & Pharmacology and Center for Interdisciplinary Research in Cardiovascular Sciences, West Virginia University, Morgantown, West Virginia, 26505, USA E-mail address: smustafa@hsc.wvu.edu

Adenosine, a purine nucleoside acts through its cell surface receptors (A1, A2A, A2B, and A3) via its coupling to G-proteins. Adenosine causes an increase in coronary flow mostly through A2A adenosine receptors. However, the involvement of other adenosine receptors in the modulation of coronary flow regulation is not well understood. Using adenosine receptor knockout mice, we have investigated the roles of each adenosine receptor subtype in the regulation of coronary flow. Recently, using the isolated mouse heart preparation from A3 adenosine receptor knockout mice, we reported an increased A2A receptor-mediated coronary flow [1]. Similarly, we found an increase in coronary flow in A1 adenosine receptor knockout mice [2]. These data support inhibitory roles for A3 and A1 adenosine receptors in the regulation of coronary flow. Additionally, in A2A adenosine receptor knockout mice, the response to A2A selective agonist (CGS-21680) was totally abolished [3]. On the other hand, adenosine A2A receptor knockout mice showed a decrease in blood flow to NECA (non-selective agonist). Currently, we are investigating the coronary flow responses to both NECA and CGS-21680 in A2B adenosine receptor knockout mice. Further, real time-PCR for all four adenosine receptor knockout mice showed no significant change in all four adenosine receptors. These data support stimulatory roles for A2A and A2B adenosine receptors in the regulation of coronary flow. These observations provide new evidence for the presence and possible roles of all four adenosine receptor subtypes in coronary flow regulation. (Supported by National Heart Lung and Blood Institute grant # HL-027339).

Subcellular Aspects of Adenosine Receptor-Mediated Signaling in Ventricular Myocardium

Robert D. Lasley

Department of Surgery, University of Kentucky College of Medicine, Lexington, KY, USA rlasley@uky.edu

For years the effects of adenosine in ventricular myocardium (intact hearts and myocytes) were attributed solely to A1 receptor subtypes. However evidence generated by numerous laboratories over the past several years indicates that these effects are due to multiple adenosine receptor subtypes and interactions with various signaling pathways. Although the adenosine A3 receptor subtype has received much interest in adenosine's effects in ischemic myocardium, there is additional evidence for the involvement of the A2a receptor subtype. We have previously reported that: 1) intracoronary infusion of the A2a agonist CGS21680 increased load-insensitive contractility in stunned, but not normal, in vivo porcine myocardium, independent of changes in coronary blood flow, [1] and 2) the preconditioning (PC) effect of the adenosine agonist AMP579 in in vivo rat myocardium is blocked by both the A1 antagonist DPCPX and the A2a antagonist ZM241385 [2]. Studies from other investigators using isolated hearts and cardiomyocytes suggesting a role for A2a receptor effects are consistent with our report that rat ventricular myocytes express A2a receptors [3]. Additional observations from our laboratory indicate that although A2a receptor activation does not induce PC, ZM241385 blocks the cardioprotective effects of the A1 agonist CCPA and the non-selective agonist NECA to a similar degree as DPCPX. Additional preliminary observations from our laboratory suggest that A2b receptors are expressed in ventricular myocardium and myocytes. However based on Western blot analysis and immunocytochemistry the majority of A2a and A2b immunoreactivity is expressed in cytosolic, not membrane, fractions. In addition to adenosine receptor subtype interactions, adenosine receptors have been reported to interact with other G protein coupled receptors (GPCR), such as opioid receptors, in various tissues via signaling cross-talk and/or receptor hetero-dimerization. Although there have been few such studies in the heart, it has been reported that in vivo A1 agonist PC is blocked by a delta opioid receptor antagonist and morphine-induced PC is blocked by DPCPX [4]. We have observed similar findings that indicate that this interaction is selective for delta, but not kappa, opioid receptors. Adenosine receptors have been shown to couple to the activation of mitogen activated protein kinases (MAPKs) in various tissues. Studies in our laboratory indicate that adenosine receptors activate MAPKs in both intact myocardium and ventricular myocytes; however this activation occurs in distinct subcellular compartments. Our observations also indicate that the acute and delayed PC effects of adenosine receptor activation are dependent on MAPK signaling. Thus there is increasing evidence that the effects of adenosine in ventricular myocardium are mediated by multiple receptor subtypes and their interactions as well as via the modulation of cardiomyocyte subcellular signaling.

Use of Adenosine Compounds for Identifying Patients Prone to Suffer from Tachyarrhythmias

Raphael Rosso, M.D.

Department of Cardiology, Tel Aviv — Sourasky Medical Center and Sackler School of Medicine, Tel Aviv University

In our presentation we review our experience in the use of adenosine-compounds for diagnostic purposes in cardiac arrhythmias. Adenosine-compounds are the treatment of choice for common supraventricular tachyarrhythmias involving the atrio-ventricular node. Moreover, adenosine-compounds can also be used for diagnostic purposes.

Injection of adenosine compounds during sinus rhythm leads to short lasting bradycardia (sinus arrest or block of conduction along the AV-node). This effect is short lasting and invariably resolves within seconds. The induced bradyarrhythmia is always followed by sinus tachycardia. This unique effects of adenosine compounds may be used for the following diagnostic purposes: 1) to reveal the presence of Wolff-Parkinson-White in patients with minor preexcitation due to a left sided location of the accessory pathway and fast AV nodal conduction, 2) to reveal the presence of dual AV node physiology in patients with AV-node reentry tachycardia, 3) to reveal the presence of concealed accessory pathways in patients with AV reentry tachycardia. The last two effects can be useful when evaluating patients who have a history of palpitations but no documented arrhythmias. Finally, since patients with long QT syndrome develop torsade de pointes during sudden deceleration (bradycardia-dependent) or sudden acceleration (tachycardia dependent torsade) of the heart rate, we have used adenosine-compounds for diagnosing long QT syndrome in patients with borderline QT intervals.

Bone and Cartilage

Integrating Local and Systemic Regulation of Bone Turnover Through P2 Receptor Signalling

J.A. Gallagher

Human Bone Cell Research Group, Department of Human Anatomy and Cell Biology, The University of Liverpool, Liverpool, L69 3GE U.K. jag1@liv.ac.uk

Bone is continually remodelled throughout life by bone-resorbing osteoclasts and bone-forming osteoblasts. This remodelling is regulated by a complex interaction of systemic hormones, including parathyroid hormone (PTH), and local factors, including mechanical loading, which together ensure that the skeleton adapts to physiological requirements. Extracellular nucleotides acting through multiple P2 receptors play a major role in integrating the local and systemic signalling. Since the early descriptions of P2 receptor expression in bone1, functional and molecular evidence has accumulated for the presence of multiple P2X and P2Y receptors on osteoblasts and osteoclasts. In addition to expressing P2 receptors, bone cells release ATP constitutively2 and this release is enhanced by mechanical stimulation. The exact mechanism for the constitutive release of ATP by osteoblasts remains to be identified, although exocytosis probably plays some part.

Some of the the functional consequences of ATP release in bone and subsequent activation of P2 receptors on bone cells have been identified. For example, activation of osteoblastic P2Y1 receptors leads to an up regulation of receptor activator of nuclear factor-kB ligand (RANKL), which stimulates the formation and activity of osteoclasts and thereby stimulates bone resorption3, and P2X7 receptors are expressed on osteoblasts and osteoclast and may play important roles in mechanotransduction and osteoclast formation4. One observation that we regard as particularly significant is the effect of activation of osteoblastic P2Y receptors on the induction of c-fos, an immediate early gene that plays a key role in the proliferation and differentiation of bone cells. Activation of osteoblastic P2Y receptors alone results in a only a moderate induction of c-fos, whereas dual activation of P2 receptors and PTH receptors leads, through multiple levels of interaction in downstream signalling, to a massive synergy in c-fos expression5. This synergy provides a molecular mechanism whereby locally released extracellular nucleotides can sensitise cells to systemic PTH and thereby activate remodelling at discrete sites in bone. This may represent a common mechanism to integrate systemic and local signals in the regulation of turnover in other tissues.

P2X7 nucleotide receptor functions in bone cells

Nattapon Panupinthu, Jasminka Korcok, Joseph T. Rogers, Stephen M. Sims and S. Jeffrey Dixon

CIHR Group in Skeletal Development and Remodeling, Schulich School of Medicine & Dentistry, The University of Western Ontario, London, Canada, N6A 5C1 E-mail: jeff.dixon@schulich.uwo.ca

Since ATP is released from cells in response to mechanical stimulation, it has been proposed that P2 nucleotide receptors mediate mechanotransduction in bone.1 Mice with targeted disruption of the P2X7 receptor gene exhibit diminished periosteal bone formation together with excessive trabecular bone resorption2 and impaired response to mechanical loading.3 Our goal was to understand the roles of P2X7 receptors in osteoblasts (bone forming cells) and osteoclasts (bone resorbing cells). Whereas the expression of functional P2X7 receptors has been established in osteoclasts,4 their existence in osteoblasts has been controversial. Osteoblast-enriched cultures were obtained from calvariae of newborn rats, and wild-type (WT) and P2X7 receptor knockout (KO) mice. Osteoblast-enriched cultures were supplemented with ascorbic acid and β-glycerophosphate to induce formation of mineralized nodules. Cells expressing functional P2X7 receptors were associated with bone nodules. Benzoylbenzoyl-ATP (BzATP, P2X7 agonist), but not UTP, promoted mineralization in rat calvarial cell cultures. Moreover, mineralization in KO cultures was markedly reduced compared to WT, consistent with a role for P2X7 receptors. Expression of osteoblast marker genes (encoding Runx2, Osterix, osteocalcin and bone sialoprotein) was assessed by real-time PCR. Transcript levels were significantly enhanced by BzATP in rat calvarial cell cultures and suppressed in cultures from KO mice compared to WT (except Runx2). Cyclooxygenase inhibitors abolished the stimulatory effect of BzATP on bone formation without affecting mineralization in untreated cultures. Thus, P2X7 receptors stimulate osteoblast differentiation through a cell-autonomous, cyclooxygenase-dependent mechanism. The lifespan of osteoclasts is an important regulator of bone resorption. Osteoclasts were isolated from long bones of newborn WT and KO mice and cell survival was assessed. In the absence of exogenous nucleotides, higher numbers of WT than KO osteoclasts were apoptotic at 6 h. Consistent with this observation, fewer WT osteoclasts survived at 12 h. Interestingly, BzATP had no additional effect on either apoptosis or survival. Control experiments revealed that WT and KO osteoclasts were equally susceptible to apoptosis induced by the proapoptotic agent staurosporine. Brilliant blue G (P2X7 antagonist) decreased apoptosis and increased survival of WT osteoclasts, providing additional evidence for involvement of the P2X7 receptor. Treatment of WT osteoclasts with alkaline phosphatase (to hydrolyze endogenous nucleotides) increased their survival to levels comparable to KO osteoclasts, consistent with activation of P2X7 receptors by constitutively released nucleotides. In summary, P2X7 receptors stimulate osteoblast differentiation and induce osteoclast apoptosis. Since the P2X7 receptor both enhances bone formation and suppresses bone resorption, it represents a potential target for the development of drugs with combined anabolic and antiresorptive effects. Supported by the Canadian Institutes of Health Research (CIHR).

P2Y nucleotide receptor function in osteoblasts

Isabel Orriss, Geoffrey Burnstock and Timothy R. Arnett

Department of Anatomy and Developmental Biology, University College London, London WC1E 6BT, UK. e-mail: t.arnett@ucl.ac.uk

Accumulating evidence suggests that extracellular nucleotides, signalling through P2 receptors, play a role in modulating bone cell function. ATP and ADP stimulate osteoclastic resorption, whilst ATP and UTP are powerful inhibitors of bone formation by osteoblasts1. We investigated changes in the expression of P2 receptors with cell differentiation in primary osteoblast cultures. Rat calvarial osteoblast, derived by trysin/collagenase digestion and cultured for up to 10 days, were loaded with the intracellular Ca2+-sensing fluorophore, Fluo-4 AM, and a fluorescence imaging plate reader (FLIPR) was used to measure responses to nucleotide agonists. Peak responses occurred within 20 sec, and were evoked by ATP or UTP at concentrations as low as 2 µM. Numbers of osteoblasts doubled between days 4 and 10 of culture but the peak intracellular Ca2+ response to ATP or UTP increased up to 6-fold over the same period, indicating that osteoblast responsiveness to nucleotides increased as cell differentiation proceeds. The approximate order of potency for the most active nucleotide agonists at day 8 of culture was ATP > UTP & ATPγS > ADP > UDP, consistent with the expression of functional P2Y2, P2X{ia2}, P2Y4, P2Y1 and P2Y6 receptors. Smaller responses were elicited by 2-MeSATP, Bz-ATP and α,β-meATP, additionally suggesting the presence of functional P2X1, P2X3, P2X5 and P2X7 receptors. Expression of mRNA for the ATP and UTP-selective P2Y2 receptor increased strongly between days 6 and 15 in primary osteoblasts, whereas mRNAs for the P2Y4 (also ATP/UTP selective) and P2Y6 (UDP/UTP selective) receptors were highly expressed at intermediate time points. In contrast, mRNA for the cell-proliferation-associated P2X5 receptor decreased to undetectable as osteoblasts matured, but mRNA for the cell death-associated P2X7 receptor was detected at all time points. Similar trends were evident using immunostaining and Western blotting for P2 receptors. Exposure to 1–10 µM ATP or UTP during days 10–14 of culture was sufficient to cause near-total blockade of alizarin redstained bone nodules formed by osteoblasts; however, UDP and ADP were without effect. Microscopy revealed that ATP and UTP-treated osteoblasts deposited abundant collagenous material with the characteristic morphology of bone nodules, but that mineralisation failed to occur. In osteoblast cultures treated with 10 µM ATP or UTP, the activity of alkaline phosphatase (thought to participate in bone mineralisation) was reduced by >50%. Our results show that there is a shift from P2X to P2Y expression during osteoblast differentiation in culture, with mature osteoblasts preferentially expressing the P2Y2 receptor and, to a lesser extent, P2Y4 and P2Y6 receptors. Taken together, these data suggest that the P2Y2 receptor and, possibly, the P2Y4 receptor could function as ‘off-switches’ for mineralised bone formation. It should be noted, however, that since several ectonucleotidases can generate pyrophosphate (PPi) from nucleotide triphosphates, and that PPi is a potent inhibitor of bone mineralisation (at ≥1 µM in our in vitro system), the possibility remains that a component of the effect of extracellular nucleotides on mineralisation could also occur independently of P2 receptors. We are grateful for the support of the Arthritis Research Campaign.

The role of nucleotides and P2 receptors in intercellular signaling in bone

Niklas Rye Jørgensen

Dep. of Clinical Biochemistry, Copenhagen University Hospital Hvidovre, Hvidovre, Denmark Email: niklas@dadlnet.dk

The regulation of bone turnover is a complex and finely tuned process. Many factors regulate bone remodelling, including hormones, growth factors, cytokines etc. However, little is known about the signals coupling bone formation to bone resorption, and how mechanical forces are translated into biological effects in bone.

Intercellular calcium waves are increases in intracellular calcium concentration in single cells, subsequently propagating to adjacent cells, and can be a possible mechanism for the coupling of bone formation to bone resorption and for the translation of mechanical forces on bone into biological signals. Intercellular calcium waves have been demonstrated to occur not only among osteoblasts in vitro, but also between osteoblasts and osteoclasts, in response to mechanical stimuli. Calcium signaling among osteoblasts can be propagated by the secretion of a nucleotide, possibly ATP, acting in an autocrine action to purinergic P2Y2 receptors on the neighboring cells, leading to IP3 generation and subsequent release of calcium from intracellular stores. In the signaling from osteoblasts to osteoclasts, paracrine action of ATP seems to be responsible, with the involvement of purinergic P2X7 pore forming receptor.

Activation of P2 receptors in bone cells not only involves the propagation of intercellular calcium waves but also the proliferation and activity of the cells. Further, alterations in receptor expression and function have effects on bone development and turnover as well as on bone quality and fracture risk in vivo.

Thus, nucleotide-mediated intercellular calcium signalling is taking place in vitro among osteoblasts and between osteoblast and osteoclasts, and might be a mechanism by which mechanical forces regulate bone turnover in vivo. The mechanisms involved in the calcium signaling are definitely important in normal bone development and turnover, and might be future targets for pharmacological manipulation of bone metabolism.

Immunology and Inflammation

Activation of P2X7 by ADP-Ribosylation

Friedrich Koch-Nolte

Institute of Immunology, University Hospital, Martinistr. 52, Hamburg, 20246 Germany nolte@uke.uni-hamburg.de

The cytolytic P2X7 purinoceptor is widely expressed by inflammatory cells. P2X7 has attracted attention because of its unique capacity to induce the formation of a large pore and cell death. Activation of P2X7 requires much higher concentrations of ATP than activation of other P2X family members. P2X7 on murine T cells can also be activated by ADP-riboslyation of cell surface proteins catalyzed by the toxin like ecto-ADP-ribosyltransferase ART2 following release of the ART substrate, NAD, from cells [1]. Much lower concentrations of NAD than ATP suffice to activate P2X7. However, at saturating doses, ATP induces stronger activation signals than ADP-ribosylation. Moreover, activation of P2X7 is readily reversed after short exposures to ATP followed by washing away of the soluble ligand, wheras short exposures to NAD followed by washing chronically activate P2X7 by providing covalently bound ligands. Using site directed mutagenesis we have identified the sites of ADP-ribosylation on P2X7. The results shed light on the issue whether P2X7 is activated by ADP-ribose moieties attached to neighboring proteins or to P2X7 itself.

Supported by DFG grant No310/6-1 to FH and FKN

Adenosine 2A Receptor (A2AR) Agonists Regulate Renal Nitric Oxide Production in Diabetic Nephropathy

Alaa S. Awad, Elizabeth Duong, Liping Huang, Joel Linden and Mark D. Okusa

Departments of Medicine, University of Virginia, Charlottesville, VA 22908 awad@virginia.edu


Nitric oxide (NO) regulates renal hemodynamics, endothelial function and inflammation. However, the role of NO in the pathophysiology of diabetic nephropathy (DN) remains controversial. We have shown previously that A2A receptor agonists prevent renal tissue injury and inflammation in DN [Am J Physiol Renal Physiol 2006, in press]. Whether the protective effect of A2A receptor agonists is mediated by regulation of renal NO production is not known. We hypothesized that A2A agonists ameliorate renal endothelial dysfunction and inflammation associated with DN through an effect on NO production.


Diabetes was induced with single iv injection of streptozotocin (STZ) in Sprague Dawley rats (50 mg/kg). Urinary NO end products (NOX) and nitric oxide synthases (NOS) mRNA were evaluated in control, diabetes with vehicle and diabetes with ATL146e, a selective A2a agonist (10 ng/kg/min) via osmotic pump over 6 weeks.


Blood glucose increased in the diabetes and treatment groups 48 hours after STZ-induction of diabetes. Renal medullary blood flow and urinary NOX excretion were reduced significantly in diabetes group (67%; p < 0.001 and 60%; p < 0.01 from control) and restored to normal levels by ATL146e (86% and 100% from control), respectively. There was a significant correlation between urinary albumin excretion rate and urinary NOX (r = 0.84; p < 0.0001). We next sought to determine renal NO synthase (NOS) isoforms altered in DN. By using realtime PCR, diabetes led to a decrease in renal eNOS mRNA expression (67% from control) that was reversed by ATL146e (137% from control; p < 0.05 from diabetes). iNOS mRNA expression was not detectable in control but was induced in diabetes, an effect that was reversed by ATL146e. Immunohistochemistry with iNOS antibody of the kidney tissue confirmed the result of mRNA expression and was corelated with increased macrophages infiltration, a major source of iNOS, in diabetic kidneys. In contrast, there was no significant change in nNOS gene expression between groups.


These results demonstrate that DN is associated with decreased renal production of NO mainly through decreased expression of eNOS an effect reversed by A2A agonists. Furthermore an increase iNOS due in part to macrophage infiltration in DN is reversed with ATL146e.


The effects of A2A agonists on ameliorating the previously described functional and morphological changes associated with DN may in part be due to effects on NO.

Adenosine in Dermal and Hepatic Fibrosis

Edwin SL Chan

New York University School of Medicine, New York, NY 10016, USA Chane01@nyu.edu

Adenosine promotes the healing of wounds and matrix deposition in the skin, actions known to be mediated through the A2A receptor. The release of adenosine also accounts for the anti-inflammatory activity of methotrexate, one of the most widely used anti-rheumatic agents. We hypothesized that ligation of the A2A receptor is responsible for the development of hepatic fibrosis, one of the most serious complications of chronic methotrexate therapy, as well as cirrhosis arising from other causes such as ethanol consumption. Epidemiological studies have demonstrated a protective effect of caffeine in hepatic cirrhosis and we speculate that this protective effect is also related to the antagonistic activity of caffeine at the adenosine receptor. The capacity of adenosine for modulating fibrous tissue development in the skin is also relevant to the pathogenesis of another rheumatic disease, scleroderma, where skin fibrosis is a prominent feature. Using in vitro and in vivo models, we have demonstrated a role for the A2A receptor in the pathogenesis of these disorders. These findings provide a strong rationale for the exploitation of adenosine receptor antagonist therapies in future treatment of fibrotic diseases.

Adenosine receptors and the regulation of chronic lung disease

Michael R. Blackburn

Department of Biochemistry and Molecular Biology, University of Texas-Houston Medical School michael.r.blackburn@uth.tmc.edu

Chronic lung diseases are associated with persistent lung inflammation and damage that lead to the progressive loss of lung function. In contrast to most injury and repair responses, the inflammation seen in these disorders is chronic and may last throughout the life of the afflicted individual. The mechanisms that are responsible for the intensity and chronicity of these disorders have not been elucidated. Adenosine is a signaling nucleoside that can elicit a variety of cellular functions by engaging cell surface adenosine receptors. Adenosine levels are elevated in the lungs of patients with chronic lung disease suggesting this signaling pathway may be involved in these disorders. My laboratory makes use of genetically modified mice to examine the importance of adenosine signaling pathways in aspects of chronic lung disease. We have generated mice deficient in the enzyme adenosine deaminase (ADA) that controls the levels of adenosine in tissues and cells. Adenosine levels accumulate in the lungs of these mice and are associated with increased lung inflammation and histopathologies seen in asthma and chronic obstructive pulmonary disease including fibrosis and alveolar airway enlargement. We have used both genetic and pharmacologic approaches to examine the contributions of the different adenosine receptors to the inflammation and damage seen in the lungs of ADA-deficient mice. In addition, we have examined the contribution of adenosine and its receptors in other models of chronic lung disease. My presentation will use data gathered in the above model systems to highlight the various pro-inflammatory/tissue destructive effects of adenosine receptors as well as anti-inflammatory/tissue protective effects, as they pertain to aspects of chronic lung disease.

ADP-ribosylation by ARTs introduces diacritic modifications in the cytokine network

Michel Seman

EA1556 University Denis Diderot-Paris7 and INSERM U519, Faculty of Medicine and Pharmacy, 22 Boulevard Gambetta, 76000, Rouen, France michel.seman@univ-rouen.fr

Mammalian mono ADP-ribosyl transferases (ARTs) constitute a family of ectoenzymes catalyzing the transfer of ADP-ribose from NAD+ onto arginine residues of target proteins. They are thus responsible for unusual posttranslational modifications of proteins in the extracellular compartment. NAD+ is released during inflammation and can be detected in inflammatory liquids. This provides enough substrate to ARTs and renders possible ADP-ribosylation of neighboring membrane targets on the surface of ART expressing cells such as integrins or P2X7, and free soluble proteins present in surrounding extracellular fluids including cytokines and growth factors. Several cytokines can thus be ADP-ribosylated by ART1 but others are not although they all contain arginine residues in their primary sequence. This provides evidence for selective ADP-ribosylation sites. Cytokine ADP-ribosylation can either block their functions and/or modulate their bioavailability. ART-mediated ADP-ribosylation of cytokines represents a new mechanism of regulation in the immune system which may contribute to the control of inflammation and host-tumor interactions. It may also participate in the resistance of patients to cytokine-based therapies.

Supported by grants from the Ministe` re de la Recherche, Ligue Nationale contre le cancer, Association pour la recherche sur le cancer.

Nicolò Copernico Award: From “Hellstrom Paradox” to Redox-Adenosinergic Rule of Immune Response

Michail V. Sitkovsky

New England Inflammation and Tissue Protection Institute, Consortium at Northeastern University, Boston, MA, USA M.Sitkovsky@neu.edu

We will describe the redox-adenosinergic mechanism of protection of normal tissues from excessive immune damage. These studies were motivated by the long-standing “Hellstrom Paradox” of the peaceful co-existence of tumors and anti-tumor immune cells in the same cancer patient. We summarize the more than 25 years of research that was focused first on mechanisms of cell-cell interactions during T-cell mediated lethal hit delivery to tumor target cells, then on mechanisms that propagate or may inhibit the T cell receptor-triggered signaling. We provide evidence, which suggests that down-regulation of overactive immune cells in inflamed areas may be triggered by excessive collateral immune damage to endothelial cells and microcirculation. The ensuing interruption of normal blood and oxygen supply leads to low tissue oxygen tension. This is associated with an accumulation of extracellular adenosine and signaling through A2A and A2B adenosine receptors on the surface of surrounding cells, including activated myeloid and T cells. This chain of events culminates in inhibition of activated immune cells in a delayed, negative feedback manner due to the well-established immunosuppressive effects of cAMP-elevating A2 adenosine receptors. We also describe the rationale behind the model that the tissue protecting effects of hypoxia involve both hypoxia-stabilized Hypoxia Inducible Factor-1α and A2 adenosine receptors. Similar mechanism mat explain Hellstrom paradox. An example of the predictive power of the redox-adenosinergic rule is provided by the confirmed exacerbation of acute lung injury by hypoxia-eliminating supplemental oxygen therapy.

Target genes of extracellular ATP in human dendritic cells

Didier Communi1, Frédéric Marteau1, Michael Horckmans1, Nathalie Bles1, Jean-Marie Boeynaems1,2

1Institute of Interdisciplinary Research, IRIBHM, Université Libre de Bruxelles, Belgium; communid@ulb.ac.be 2 Department of Medical Chemistry, Erasme Hospital, Université Libre de Bruxelles, Belgium

Extracellular ATP affects the maturation of human monocyte-derived dendritic cells (MoDCs), mainly by inhibiting Th1 cytokines, promoting Th2 cytokines, and modulating the expression of costimulatory molecules. Comparison of ATP and PGE2 expression profiles in human moDCs revealed an extensive number of target genes involved in immunosuppression, maturation and chemotaxis.

More particularly, ATP elicited a drastic up-regulation of two mediators involved in immunosuppression: thrombospondin-1 and indoleamine 2,3-dioxygenase. Extracellular ATP released from damaged cells and previously considered as danger signal is thus a potent regulator of mediators playing key roles in immune tolerance.

We also observed that ATP is able to regulate several chemokines and chemokine receptors. More particularly, adenine nucleotides induced down-regulation of major monocyte recruiters such as MCP-1 and MIP-1α chemokines and reduced the capacity of MoDCs to attract monocytes and immature DCs.

Pharmacological data support the involvement of the P2Y11 receptor in the regulation of these ATP target genes in human MoDCs. Consequently nucleotides derivatives may be considered as useful tools for DC-based immunotherapies. Several other promising genes regulated by ATP have been identified in human DCs and their study is ongoing.

P2 Receptors: Role in host defence against intracellular microorganisms

Robson Coutinho-Silva1, Suzana Passos Chaves1, Camila Marques da Silva1,2, Pedro M. Persechini1, Bartira B Bergmann1 and David Ojcius3

1Instituto de Biofísica Carlos Chagas Filho and 2Dept. Ciências Morfológicas — Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brasil; 3Division of Natural Sciences, University of California, Merced, CA rcsilva@biof.ufrj.br

A growing number of studies have demonstrated the importance of ATPe-signalling via P2 receptors as an important component of the inflammatory response to infection. More recent studies have shown that ATPe can also have a direct effect on infection by intracellular pathogens, by modulating membrane trafficking in cells that contain vacuoles that harbor intracellular pathogens, such as mycobacteria and chlamydiae. A conserved mechanism appears to be involved in controlling infection by both of these pathogens, as a role for phospholipase D in inducing fusion between lysosomes and the vacuoles has been demonstrated. P2X7 receptors is involved in clearance of intracellular bacteria, but in mycobacteria, infection seems to controlled by an additional P2 receptor. Other P2-dependent mechanisms are most likely operative in the cases of pathogens such as Leishmania, which survive in an acidic phagolysosomal-like compartment. We explored the possibility that P2 receptors could be used as a therapeutic drug against Leishmania infection. BALB/c mice infected with L. amazonensis were treated for three weeks, twice a week, with PBS, ATP (50 mM), or UTP (5 mM) in 20 2l injections in foot lesions. We observed that ATP and UTP reduced the parasitic load by two-log units compared to control mice treated with PBS.. In vitro, peritoneal macrophages were infected for 48 h at 37°C, and then treated with ATP (500 µM), ADP (100 µM), adenosine (100 µM), UTP (100 µM), UDP (100 µM) or oATP (300 µM) in PBS for 30 min. Only ATP showed significant anti-amastigotic activity (36% inhibition), with no significant cytotoxicity observed for any of the nucleotides, and UTP and UDP inducing production of nitric oxide. Measurements of calcium flux and apoptosis induced by the nucleotides data suggest possible modulation of at least two P2 receptor during in vitro infection with L. amazonensis, one being sensitive to ATP and the other, to UTP. These results suggest that P2 receptors are potential targets for anti-leishmanial therapy, that the immunological environment is probably important for this activity, and that there are several P2 receptors involved in the anti-microbial effect of the nucleotides.

Acknowledgment FAPERJ and CNPq.

Polymorphisms in the P2X7 receptor and their clinical associations

James S. Wiley,1 Suran Fernando,2 Kristy Skarratt,1 Bernadette Saunders,2 Ronald Sluyter,1 Ben Gu,1 Stephen Fuller1, Anne Shemon,1 Phuong Dao-Ung,1 Jenny Georgiou,1 and Warwick Britton2

1Department of Medicine, University of Sydney, Nepean Hospital, Penrith NSW 2750, and 2Centenary Institute of Cancer Medicine and Cell Biology, Camperdown, 2042, Australia Email: wileyj@medicine.usyd.edu.au

The search for the function of the P2X7 receptor has drawn on results from two models; the gene-deleted mouse and rare genetic variants within the human population. Studies of the knock-out mouse by both the Pfizer and Glaxo groups have pointed to P2X7 as a pro-inflammatory gene which may have a role in bone structure. Parallel clinical studies by our group and others have suggested that P2X7 may also have a role in innate immunity and resistance to certain infections.

The innate immune system is vital for the control of infection by intracellular pathogens such as mycobacteria, chlamydia and toxoplasma. These organisms evade the adaptive immune response by entering and replicating inside host cells and tissues. Recent studies have shown that activation of the P2X7 receptor on cells of the monocyte — macrophage series is needed for killing of intracellular mycobacteria and that the same activation signal leads to subsequent apoptotic death of the infected macrophage. Wide variations in P2X7 function have been observed between individuals and genetic factors play a major role in this variability. Much of the variability is due to the polymorphic structure of the gene with an average of one single nucleotide polymorphism (SNP) every 160 bp of coding DNA and we have described at least 5 SNPs causing loss of function and a further one has been reported giving gain of function.

In around 30% of the Caucasian population, a Glu-496 to Ala polymorphism in P2X7 receptors (1513 A→C), of the carboxyl terminus leads to loss of function. Homozygosity for this polymorphism abolishes ATP-induced channel dilatation (pore formation) while the heterozygous state gives cells with half the pore function of cells with wild type P2X7 protein. A second polymorphism, Ile-568 to Asn (1729 T→A) lies in a trafficking motif of the carboxyl terminus and prevents normal trafficking and surface expression of the receptor. A third polymorphism, Arg-307 to uncharged glutamine (946 G→A) lies within the ATP binding pocket of the extracellular domain and abolishes the binding of ATP to the receptor with severe functional loss. A fourth polymorphism occurs as a splice site mutation at position +1 of the first intron of the gene (151+1 g→t) and gives rise to a null allele in 1–2% of the Caucasian population. While each of the latter three polymorphisms (1729A, 946A and 151+1t) occurs at an allele frequency of 1–2% in Caucasians, none have been found in a large cohort of S.E. Asian immigrants to Australia.

The effect of loss-of-function SNPs appears to be additive such that macrophages from compound heterozygous subjects have total ablation of P2X7 function. Thus we have observed severe illness due to Toxoplasma gondii in two compound heterozygotes with total absence of P2X7 function. In two large case control cohorts we have studied the reactivation of tuberculosis in immigrants of mainly S.E. Asian origin after arrival in Australia. The 1513 A→C polymorphism was strongly associated with extrapulmonary but not pulmonary tuberculosis in both the first (OR 3.99, P< 0.001) and second cohorts (OR 2.99, P < 0.01). These data confirm the major effects of P2X7 polymorphisms on macrophage function and innate immunity of man.

Tissue-Derived Adenosine promotes Peripheral T cell Tolerance by Stimulating the A2A Receptor

Jonathan D. Powell* +Paul E. Zarek*, Charles Drake+, Drew M. Pardoll+

*Dept. of Pharmacology, + Dept of Oncology, The Johns Hopkins University, School of Medicine, Baltimore, MD, USA 21231 poweljo@jhmi.edu

A2a receptor (A2aR) engagement by endogenous adenosine has been implicated as an important negative regulator of inflammation. Using microarray analysis, we determined that the A2aR is differentially upregulated on T cells under conditions that induce T cell tolerance. In vitro, A2aR WT, but not A2aR null T cells that were initially activated in the presence of CGS-21680 (CGS) were hyporesponsive upon rechallenge; they were anergic. In vivo, the adoptive transfer of A2aR null HA-specific 6.5 T cells into mice expressing HA as a self-antigen led to rapid autoimmune-induced death when compared with the transfer of T cells from WT mice. Furthermore, treating mice with just 4 days of CGS inhibited autoimmune-induced death and promoted the induction of T cell anergy. In this model, mice that survive the initial adoptive transfer develop inducible LAG-3+ regulatory T cells that protect the mice from a subsequent rechallenge with clonotypic T cells. LAG-3 expression levels and kinetics are enhanced in A2aR WT cells from CGS-treated mice, but decreased in A2aR null T cells, regardless of CGS-treatement. CGS-treated mice that survived the initial adoptive transfer were protected against a subsequent rechallenge with lethal numbers of clonotypic cells. Conversely, CGS failed to protect mice from the adoptive transfer of LAG-3 null T cells. It is notable that the concentration of adenosine in the tumor mciroenvironment is in the mM range. This suggests that one mechanism by which tumors evade immune destruction is via A2aR-induced tumor-antigen-induced tolerance. To test this hypothesis we examined the efficacy of tumor vaccines in A2aR null mice or when given in conjunction with A2aR antagonists. We found that in the A2aR null mice or in mice which received A2aR antagonists, the efficacy of immunotherapy was greatly enhanced. Overall, our data implicate tissue derived adenosine as a novel mechanism for promoting peripheral T cell tolerance.

Round table on Caffeine and Health

Caffeine and Headache: An Update

Astrid Nehlig

INSERM U 666, Strasbourg, France e-mail: nehlig@neurochem.u-strasbg.fr

The use of caffeine as an adjunctive constituent of analgesic medications dates back to 1875 while the first clinical trials were performed after 1950. The combination of caffeine with current prescription and nonprescription drugs in alleviating migraine and tension-type headache-related symptoms and pain has been the subject of a number of studies.

Caffeine has intrinsic antinociceptive actions in animal models. These are linked to its antagonism at the level of adenosine receptors. Adenosine A1 receptor activation is antinociceptive by decreasing and A2 receptor activation is pronociceptive by increasing cAMP levels in the sensory nerve terminals. Adenosine A3 receptor activation produces pain behaviors due to the release of histamine and serotonin from mast cells and subsequent actions on nerve cell terminals. The intrinsic antinociceptive actions of caffeine have been proposed to result from actions at supraspinal sites and may involve inhibition of presynaptic adenosine receptors on cholinergic nerve terminals. In addition, during headaches, adenosine receptor-mediated vasodilation and/or irregular vascular tone contribute to pain.

Based on a pooled analysis of 30 clinical studies including about 10,000 patients suffering from various types of pain including headaches, the combination of caffeine with current analgesics was reported to increase the analgesic effect of aspirin or acetaminophen by 41%. Caffeine is well known to alleviate withdrawal-induced headache which occurs after the abrupt cessation of caffeine intake. This effect is strongly linked to the psychological satisfaction related to the ingestion of caffeine; this is especially true for the first cup of coffee in the morning. In tension-type headaches, the efficacy of the combination of caffeine with acetaminophen, aspirin or ibuprofen has been repeatedly reported both for pain intensity and total pain relief. In this type of headache, caffeine displays also analgesic properties when given alone. In migraine, the combination of acetaminophen, aspirin and caffeine is more effective than the placebo in alleviating pain and associated symptoms even in severely disabled migraineurs. However, in the latter type of headache, the studies do not allow to outline the specific role of caffeine in the drug combination tested since no comparison was performed with combinations containing or not caffeine or with caffeine alone. A recent work on migraine suggested to use a combination of products that would act on a whole subset of physiological regulations that are impaired during migraine attacks.

The role of cerebral blood flow changes in headaches is unclear, except in caffeine withdrawal-induced headaches. In the latter ones, cerebral blood flow velocities are increased and they are significantly decreased within 30 min after caffeine intake, suggesting that increased blood volume may be involved in caffeine withdrawal headache. In tension-type headache, there does not seem to be vascular changes related to the attack and for migraine attacks, the literature is rather in favor of a decrease in cerebral blood flow during the attacks. Thus, in these two latter headache types, caffeine does not seem to exert analgesia via its vasoconstrictive properties.

Caffeine effects on blood pressure and stress responses in persons at risk for developing hypertension

William R. Lovallo

University of Oklahoma and Veterans Affairs Medical Center, Oklahoma City, Oklahoma, USA bill@mindbody1.org

Caffeine is the worlds most widely consumed pharmacologically active substance. A constituent in coffee, tea, and soft drinks, caffeine acts to block adenosine receptors, and it is capable of interacting with all levels of the stress axis [1]. This action has implications for a range of systemic functions. Our work has concentrated on the effect of caffeine on blood pressure regulation and cortisol secretion, with an emphasis on these responses in hypertension. At the blood vessel wall, caffeine blocks A2a receptors on the prejunctional sympathetic nerve ending, reducing the ability of adenosine to regulate release of norepinephrine onto alpha adrenoreceptors on the smooth muscle cell. This increases contractile state of vascular smooth muscle and causes a rise in peripheral vascular resistance, providing a physiological basis for caffeine's ability to elevate blood pressure. This receptor blockade causes a greater and longer lasting rise in vascular resistance and blood pressures in persons with above-normal pressures [2,3]. Caffeine also increases the secretion of the stress hormone cortisol, and this effect is also greater in persons at high risk of developing hypertension. The effect of caffeine on blood pressure and cortisol secretion is not affected by the amount of caffeine the person normally consumes in their diet, raising the question of the extent of tolerance development in regular consumers. Our study of tolerance to caffeine shows that most persons do not develop a complete tolerance to caffeine's effect on blood pressure or cortisol output. Examination of individual differences in tolerance shows that half of the persons tested did not show tolerance to a morning dose of 250 mg of caffeine given in even when given a structured diet of 600 mg of caffeine per day (equivalent to 6 cups of brewed coffee) for the previous five days. These effects call into question the long-term effects of caffeine on blood pressure regulation in the development and progression of hypertension.

The effects of caffeine on anxiety- and mood-related behaviors in mice

Malika El Yacoubi, Jean Costentin, Jean-Marie Vaugeois

FRE 2735 CNRS, IFRMP 23, U.F.R. de Medecine & Pharmacie, 22 Boulevard Gambetta, 76183 Rouen Cedex, France

malika.el-yacoubi@univ-rouen.fr; jean-marie.vaugeois@univ-rouen.fr

It is now well established that the effects exerted in the brain by caffeine depend on its ability to act as an antagonist at A1 and A2a receptors of the ubiquitous neuromodulator adenosine. Caffeine can cause anxiety symptoms, especially in individuals with pre-existing anxiety disorders. It has been claimed too that caffeine use may be also associated with symptoms of depression, supporting a self-medication theory. However the precise mechanism of action of the drug is unclear. The elevated plus-maze (EPM) and the light/dark box are established behavioral tests useful to screen putative anxiogenic/anxiolytic effects of drugs. We have previously [1] investigated whether the anxiogenic effects of caffeine observed in mice depend on the blockade of A2A receptor. Caffeine acutely administered (50 or 100 mg/kg IP) induced anxiety-like effects in both procedures. Its chronic administration (50 mg/kg IP twice daily) for 1 week or consumption in the drinking water (0.3 g/l) for 8 days or 2 months were also anxiogenic in the EPM. The A2A receptor antagonists ZM241385 (up to 60 mg/kg IP) and SCH58261 (up to 10 mg/kg IP) were devoid of acute effects in both tests. One week administration of ZM241385 (30 mg/kg IP) or SCH58261 (3 mg/kg IP) had no effects in the EPM. An antagonist (DPCPX) and an agonist (CPA) at A1 receptors had no acute effects on anxiety-related indices in our hands but anxiolytic effects of adenosine acting through A1 receptors or of other A1 receptor agonists were found in other experimental conditions. An A2A receptor agonist (CGS 21680) displayed non-specific motor effects in the EPM. Acute administration of caffeine (50 mg/kg IP) induced no clear-cut anxiety-like effects in the EPM in A2a receptor knockout mice that exhibited higher basal anxiety levels than wild-type mice. Chronic administration (50 mg/kg IP twice daily) for 1 week elicited less anxiety-like behavior in A2a receptor knockout than in wild-type mice. Thus, adaptive mechanisms following mutation in A2A receptors or their long-term blockade after chronic ingestion of caffeine may be responsible for increased proneness to anxiety. However, the short-term anxiety-like effect of caffeine in mice may not be related solely to the blockade of adenosine A2A receptors, since A2a selective antagonists do not share it. It may likely be dependent on blockade of A1 receptors, in accordance with the increased anxiety-like behavior seen in A1 receptor knockout mice [2]. Adenosine and its analogues have been shown to induce “behavioral despair” in animal models believed to be relevant to depression. Selective adenosine A2a receptor antagonists (e.g., SCH 58261, ZM241385, and KW6002) or genetic inactivation of the receptor were effective in reversing signs of behavioral despair in the tail suspension test (TST) and forced swim test (FST) [3]. A2a antagonists were active in the TST using either mice previously screened for having high immobility scores or mice that were selectively bred for their spontaneous “helplessness” in this test. We hypothesized that the antidepressant-like effect of selective A2a antagonists is linked to an interaction with dopaminergic transmission, possibly in the frontal cortex. However this view of antidepressant-like effects induced by an antagonism at A2a receptors has been recently challenged.Caffeine was effective in the FST at stimulant doses but a clear-cut antidepressant-like effect could not be ascribed to it. In conclusion, data from mice show that caffeine i) can induce anxiety at high doses through A1 and possibly A2a receptors and ii) cannot be viewed as a typical antidepressant.

Respiratory System

Adenosine: A promising clinical indicator of inflammation in asthma

Riccardo Polosa

Dipartimento di Medicina Interna e Specialistica Ascoli-Tomaselli Hospitals, University of Catania, Italy polosa@unict.it

Interest in the role of adenosine in asthma has escalated considerably since the early observation of its powerful bronchoconstrictor effects in asthmatic but not normal airways. A growing body of evidence has emerged in support of a pro-inflammatory and immunomodulatory role for the purine nucleoside adenosine in the pathogenic mechanisms of chronic inflammatory disorders of the airways such as asthma. The fact that adenosine enhances mast cell allergen-dependent activation, that elevated levels of adenosine are present in chronically inflamed airways, and that adenosine given by inhalation cause dose-dependent bronchoconstriction in subjects with asthmaemphasizes the importance of adenosine in the initiation, persistence and progression of these common inflammatory disorders of the airways. These distinctive features of adenosine have been recently exploited in the clinical and research setting to identify innovative diagnostic applications for asthma. In addition, because adenosine exerts its multiple biological activities by interacting with four adenosine receptor subtypes, selective activation or blockade of these receptors may lead to the development of novel therapies for asthma.

Coordinated Nucleotide Metabolism and Vascular Leak

Sean P. Colgan1, Linda F. Thompson2, and Holger K. Eltzschig3

1Center for Experimental Therapeutics, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA

2Immunobiology and Cancer Research Program Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma

3Department of Anesthesiology and Intensive Care Medicine, Tübingen University Hospital, D-72076 Tübingen, Germany


At sites of ongoing inflammation, polymorphonuclear leukocytes (PMN, neutrophils) migrate across barrier cell types including vascular endothelia and mucosal epithelia. Such transmigration has the potential to disturb barrier properties and can result in organ dysfunction. It is recently appreciated that endogenous pathways exist to dampen barrier disruption during transmigration and may provide an important anti-inflammatory link. For example, during transmigration, PMN-derived adenine nucleotides activate endothelial and epithelial cells and induce a cyclic AMP-dependent re-sealing of barrier cell types. In recent work, we have sought to understand the link between extracellular nucleotide metabolism and vascular barrier function. In particular, these studies have revealed that activated PMN release ATP, which through the action of extracellular nucleotidases is enzymatically cleaved to adenosine. Once liberated, adenosine binds surface A2 receptors to coordinate barrier protection. As part of this analysis, we have recently addressed inflammatory microenvironmental influences, such as hypoxia, on adenine nucleotide metabolism. These studies have identified an hypoxia-dependent, transcriptionally-mediated program which coordinates the elevation of extracellular adenosine. Target molecules in this program include hypoxia-inducible factor (HIF)-dependent induction of ecto-5′-nucleotidase (CD73) and adenosine A2b receptor (A2bR), as well as HIF-dependent repression of the dipyridamole-sensitive equilibrative nucleoside transporters (ENT1 and ENT2)1, 2, 3. These studies provide new insight and potential therapeutic targets for diseases in which tissue permeability may be abnormal.

Inflammatory role of adenosine in human lung cells

Dewan Zeng, Hongyan Zhong, and Luiz Belardinelli

Department of Pharmacological Sciences, CV Therapeutics, Inc., Palo Alto, California 94304 dewan.zeng@cvt.com

It has been suggested that adenosine might have both pro- and anti-inflammatory actions depending on the pathophysiology of the diseases. In the lungs of asthmatics, inhaled adenosine or adenosine monophosphate (AMP) increases the release of histamine, tryptase and leukotrienes, suggesting a pro-inflammatory role of adenosine presumably due to the activation of adenosine receptors present in mast cells. The objective of our study was to determine the potential role of adenosine and its receptors in the human lung cells such as bronchial epithelial cells, airway smooth muscle cells and lung fibroblasts. Among the four subtypes of adenosine receptors (A1, A2a, A2b and A3), the A2b receptor is expressed at the highest level in all three types of lung cells. In airway smooth muscle cells, activation of A2b receptors leads to the release of IL-6 and MCP-1. Similarly, in lung fibroblasts, activation of A2b receptors increases the release of IL-6, which in turn promotes the differentiations of fibroblasts into myofibroblasts. In bronchial epithelial cells, activation of A2B receptors increases the release of IL-19, which leads to the release of TNF-α from a monocytic cell line. Altogether, these results suggest that activation of A2b receptors in the human lung cells increases the release of several pro-inflammatory cytokines and chemokines, which could lead to deleterious inflammatory responses in the lung.

Role of adenosine receptors in regulation of pro-inflammatory and pro-angiogenic cytokines in mast cells

Igor Feoktistov and Italo Biaggioni

Vanderbilt University, Nashville, Tennessee 37232, USA igor.feoktistov@vanderbilt.edu

Asthma is a chronic inflammatory disorder often accompanied by airway and vascular remodeling. Many of the pathologic features of asthma are attributed to the effects of mast cell-derived mediators. Adenosine is a powerful bronchoconstrictor of asthmatic, but not normal, airways. Growing lines of evidence, from early studies in isolated mast cells to more recent investigations in mice models, support a role of adenosine in Th2 and airway remodeling responses. Our own studies in this area suggested that the asthmatic response to adenosine is primarily mediated via the A2b receptors, which is selectively antagonized by enprofylline1. This work also provided evidence that adenosine is capable to regulate cytokine release from human mast cells. In recent studies, we showed that adenosine stimulates mast-dependent angiogenesis via cooperative interaction between A2b and A3 adenosine receptors. Activation of mast cell A2b adenosine receptors stimulates the synthesis of the pro-angiogenic factors VEGF and IL-8, whereas activation of A3 receptor induces the expression of angiopoietin-2.2 Thus, adenosine can contribute to the vascular remodeling associated with asthma. In addition, adenosine can contribute to Th2 response through A2b-mediated release of IL-4 and IL-13 from mast cells. The induction of IgE synthesis by the interaction between adenosine-stimulated mast cells and B-lymphocytes suggests that this mechanism is involved in the amplification of the allergic inflammatory responses associated with asthma3. The molecular mechanisms underlying this unique role of A2b adenosine receptors in promoting Th2 inflammation were explored in HMC-1 human mast cells. A2b receptors are coupled to Gs proteins linked to increase in cAMP, and are coupled to Gq to stimulate phospholipase Cβ. We found that A2b receptors upregulate IL-4 through Gq signaling that is potentiated via crosstalk with Gs-coupled pathways. These data explain the presence and underscore the importance of dual coupling of A2b receptors to Gs/Gq proteins with concurrent stimulation of diverse intracellular pathways for adenosine-dependent regulation of Th2 cytokines in human mast cells. These studies provide new insight and identify A2b receptors as a potential therapeutic target for asthma.

Nervous System

Extracellular ATP mediates rapid retinal ganglion cell injury induced by intraocular pressure spikes

Valentina RestA1, Elena Novelli1, Giovanni Vozzi2, Cristiano Scarpa2, Matteo Caleo1, Arti Ahluwalia2, Anna Solini3, Eleonora Santini3, Vincenzo Parisi4, Francesco Di Virgilio5, Lucia Galli-Resta1

1Istituto di Neuroscienze CNR — Pisa, Italy,

2Centro E. Piaggio — Dept of. Electronic Engineering — University of Pisa, Italy,

3Dept. Internal Medicine, University of Pisa, Italy,

4G.B. Bietti Eye Foundation ONLUS and Eye Clinic, University of Rome Tor Vergata, Italy, 5 Dept. Experimental and Diagnostic Medicine, Section General Pathology and Interdisciplinary Center for the Study of Inflammation (ICSI) — University of Ferrara, Italy galli@in.cnr.it

Increased intraocular pressure (IOP) may lead to retinal ganglion cell (RGC) injury, and consequent visual deficits, as it is observed in glaucoma, a leading cause of blindness. Decades of studies have shown that chronic IOP alterations cause a number of detrimental effects, including mechanical damage of RGC axons, reduced blood supply, microglia activation and release of cytotoxic factors. Much less is known of the acute effects of pressure increments, mostly for the lack of adequate tools for this investigation.

Using a novel pressure incubator we have addressed this issue in isolated rat retinas, then extended our observations in vivo.

Imaging individual RGCs before and after pressure application we found that increasing pressure to 50 mmHg for 1 min affects almost 30% of the RGCs within 1 hour. The percentage of damaged RGCs and the extent of cell injury increase when multiple pressure spikes are applied and pressure peaks are increased. However, even in the worst conditions, RGC damage is prevented by degrading extracellular ATP (eATP) with apyrase, or by blocking the P2X receptors with either oxidized ATP (300 uM), or Brilliant Blue G (0.5 uM).

In vivo, short IOP transients increase the levels of eATP in the eye fluids, damage RGCs within 1 hour, and impair RGC light-responses. Reducing eATP levels in the eye prevents RGC damage and shortens the impairment of light responses due to pressure.

These data show that pressure transients can injure RGCs within hours and demonstrates a causal link between eATP and such pressure-induced RGC damage. These findings may open the way to novel neuroprotective strategies in pathologies associated to high-pressure transients in the eye.

P2 receptors in glial cells

Maria P. Abbracchio

Laboratory of Molecular and Cellular Pharmacology of Purinergic Transmission, Department of Pharmacological Sciences, University of Milan, Italy mariapia.abbracchio@unimi.it

All types of glial cells express P2 receptors which participate to nervous system function under both physiological and pathological conditions. In the CNS, ATP has been proposed as a main signaling molecule involved in shortterm, calcium-dependent cell-to-cell signaling [1, 2]. ATP released from astrocytes activates P2 receptors on both astrocytes and other brain cells and generates a propagating wave of intracellular calcium increases allowing a homotypic and heterotypic signalling which also involves microglia, neurons and oligodendrocytes. Multiple P2X and P2Y receptors are expressed by both astrocytes and microglia [3, 4, 5]: however, these receptors are differentially recruited by nucleotides under specific patho-physiological conditions. P2Y1,2,4 and, maybe, P2X7 receptors, seem to be specifically recruited during calcium-dependent short-term signaling. Multiple P2 receptor subtypes (i.e., P2Y1,2,6,12,13 and P2X7) seem instead to cooperate to the long-term changes of astrocytes during inflammatory reactive gliosis. In these cells, P2 receptor-mediated gliosis occurs via activation of the ERK and protein kinase B/Akt pathways, and involves induction of both inflammatory and anti-inflammatory genes, cyclins, growth factors, adhesion and anti-apoptotic molecules. Conversely, genes involved in apoptotic cell death are down-regulated (for review, see: 6). Such a pattern of gene induction is consistent with the activation of neuroprotective mechanisms. Functional studies indeed suggest that, globally, the net effect of nucleotide-induced astrogliosis may be beneficial [6]. In microglial cells, exposure to inflammatory and immunological stimuli results in differential functional changes of distinct P2 receptor subtypes, suggesting that these receptors may play highly specific roles in the acquisition of the activated microglial phenotype [5, 6]. On this basis, it is currently believed that nucleotide-induced activation of both astrocytes and microglia may originally start as an acute defense mechanism aimed at protecting neurons from cytotoxic and ischemic insults. Dysregulation of this process in chronic inflammatory diseases [7] eventually results in neuronal cell damage and loss. On this basis, the full elucidation of the specific roles of P2 receptors in these effects may help exploiting the beneficial neuroprotective features of activated glial cells while attenuating their harmful properties, and thus provide the basis for novel neuroprotective strategies which specifically target the purinergic system.

Partially supported by The Italian Ministry of University and Research, progetti FIRB2001 and FIRB2003, COFIN-MIUR2004 to MPA.

Presynaptic P2 receptors — interaction with other receptors and role in neuroprotection

Lisiane O Porciúncula, Ricardo J Rodrigues, Paula M Canas, Teresa Almeida, Carla G. Silva, Jean P. Oses, A. Patrícia Simões, Catarina R Oliveira, Rodrigo A Cunha

Ctr. Neuroscience Coimbra, Fac.Medicine, Univ.Coimbra, Portugal racunha@ci.uc.pt

In the brain, ATP can either act as a neurotransmitter or neuromodulator or be involved in the communication with astrocytes or microglia. In hippocampal preparations, ATP is released in a calcium- and frequencydependent manner but its intensity-dependent release differs from that of classical transmitters (glutamate or acetylcholine, ACh). Also, ATP release was sensitive to L-type calcium channel blockers, in contrast to glutamate or ACh release. Thus, the pattern of ATP release seems to differ from that of classical neurotransmitters.

The investigation of the sub-synaptic localization of P2 receptors (P2Rs) revealed a robust immunoreactivity for P2X1,2,3Rs and P2Y1,2,4Rs in the presynaptic active zone of hippocampal terminals. Combined immunocytochemical and pharmacological characterization of glutamatergic terminals indicated a biphasic control of glutamate release by facilitatory P2X1–3Rs and inhibitory P2Y1,2,4Rs in hippocampal terminals.

Not only P2Rs but also nicotinic ACh receptors (nAChRs), namely α3*, but not α4*, were enriched in hippocampal terminals. P2X ligands displaced binding to nAChRs and nAChR ligands displaced binding to P2Rs. Glutamate or noradrenaline release from hippocampal synaptosomes was enhanced by epibatidine (102=300 nM, α-conotonin MII-sensitive) or β,γ-imido ATP (3–60 µM, prevented by NF-023 but not reactive blue-2) and, again, P2R antagonists prevented nAChR responses and, conversely, nAChR antagonists prevented P2XR responses. Also, P2X2Rs co-immunoprecipitated with α3, but not α4 subunits, showing that P2X1–3 and α3β2 nAChRs are physically linked and interact to control transmitter release.

Finally, we tested the role of P2Rs in the control of synaptotoxicity that is thought to predates neurotoxicity in neurodegenerative disorders. We tested if P2Rs controlled damage in cultured hippocampal neurons using two different noxious stimuli: 1) glutamate (100 µM for 25 min), mimicking a general feature of excitotoxicity, causing 26 ± 4% neuronal death 24 h after, and 2) Aβ1–42 (500 nM, 48 h), a causative factor of Alzheimer's disease, causing 22 ± 3% neuronal death, which displayed apoptotic features (nuclear condensation, cytochrome c release, caspase 3 activation). This neurotoxicity was preceded by a synaptotoxicity (decreased synaptophysin immunoreactivity after 12 h, when no neuronal death was observed) and dendritic atrophy (MAP-2 staining). Blockade of P2Rs (10 µM PPADS) or P2YRs (10 µM reactive blue 2), but not P2X1–3Rs (10 µM NF023 or TNP-ATP), prevented these neurotoxicity features. Notably, the selective P2Y1R antagonist MRS2179 (10 µM) was also neuroprotective. Furthermore, P2Y1R immunoreactivity was enhanced in hippocampal terminals 15 days after Aβ1–42 (2 nmol, icv) administration, at a time where mice became amnesic (passive avoidance and Y maze tests) and displayed a loss of nerve terminals (loss of synaptophysin or SNAP-25 immunoreactivities) without neuronal loss (FluoroJade C staining). This indicates that extracellular ATP may be involved in the development of the synaptotoxicity and prompts the hypothesis that P2Y1R antagonists may be potential novel neuroprotective drugs. (Supported by Portuguese Society of Neuroscience, Pfizer and FCT).

Purinergic coordination of synaptic cross-talk mediated by astrocytes

Philip G. Haydon

Department of Neuroscience, Conte Center for Integration at the Tripartite Synapse, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA pghaydon@mail.med.upenn.edu

Though it is well known that there is a tonic accumulation of adenosine in the hippocampus which results in a suppression of synaptic transmission, there is little information regarding its source and control. Using inducible transgenic animals in which we used the tetracycline-regulated gene expression system to control the expression of transgenes selectively in astrocytes, we demonstrate that astrocytes control extracellular adenosine levels through the release of ATP(Pascual et al., 2005).

During the past decade it has become appreciated that astrocytes, the predominant glial cell type of the central nervous system, respond to neuronal activity with Ca2+ oscillations which induce the release of gliotransmitters that include glutamate, ATP and D-serine (Volterra and Meldolesi, 2005;Haydon, 2001). Our recent work has focused on the roles of these glial feedback pathways in the regulation of integration in the nervous system. Using a combination of imaging, electrophysiology and molecular genetics our studies indicate that astrocytes provide balanced excitation and inhibition to neurons through immediate actions of glutamate followed by delayed actions of accumulating adenosine. These gliotransmitters serve several functions that include the synchronization of neuronal activity, the suppression of synaptic transmission, as well as in mediating adenosine-dependent heterosynaptic depression (Pascual et al., 2005).

We generated inducible astrocyte specific transgenic animals which perturbed the release of ATP from these glial cells (Pascual et al., 2005). As a consequence we were able to identify critical roles for glial-derived purines in the regulation of synaptic transmission. We demonstrate that after releasing ATP this purine is hydrolysed to adenosine where it tonically suppresses excitatory synaptic transmission. As a consequence the dynamic range for synaptic plasticity (LTP) is enhanced. Additionally, in response to high frequency activity of presynaptic afferents, we have demonstrated that astrocytes releases ATP and cause a heterosynaptic suppression of neighboring unstimulated synapses that is also mediated through the actions of adenosine. Together these results allow us to conclude that astrocyte-derived purines coordinate synaptic networks by tonically regulate the strength of synaptic transmission, the range available for synaptic plasticity and heterosynaptic depression.

Purinergic signaling in regulating myelination by action potentials

R. Douglas Fields

Nervous System Development and Plasticity Section, National Institutes of Health, NICHD, Bethesda, MD, USA fieldsd@mail.nih.gov

Myelin is the membrane wrapping around axons that provides electrical insulation essential for rapid impulse conduction. Impulse activity can affect the formation of myelin, and this may be important in regulating nervous system structure and function according to environmental experience. Our studies performed on mouse dorsal root ganglion (DRG) neurons in cell culture show that impulse activity releases ATP from DRG axons, and that this is detected by different purinergic receptors on the two types of glia forming myelin in the PNS (Schwann cells) and CNS (oligodendrocytes). Both P1 and P2 receptors are involved in axon signaling to both types of myelinating glia, but ATP is of primary importance in regulating early development and myelination by Schwann cells, where it inhibits myelination; whereas adenosine is of primary importance in regulating early development of oligodendrocyte progenitor cells (OPCs), where it stimulates myelination. After OPCs mature, activitydependent release of ATP stimulates myelin formation through a different mechanism involving release of the cytokine LIF from astrocytes, which are non-myelin forming glia of the CNS. The activity-dependent release of ATP from axons provides a mechanism for regulating myelination in the central and peripheral nervous system, through both P1 and P2 signaling at different stages of development and in different types of glia.

Cell Proliferation and Cancer

ATP: A new treatment modality in cancer?

P.C. Dagnelie

Department of Epidemiology, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Email: Dagnelie@epid.unimaas.nl

Cancer is a devastating disease. Despite decades of research, the prognosis of many cancer types is still disappointingly poor. Adenosine 5′-triphosphate (ATP), given as intravenous infusions, may have several potential applications in cancer:
  1. 1.

    Inhibition of the cancer cachexia (weight loss) syndrome, improving fatigue and quality-of-life;

  2. 2.

    Attenuation of side effects of curative anti-cancer treatment;

  3. 3.

    Anti-tumour efficacy, either alone or in combination with current cancer treatments.


In this presentation, these three effects will be outlined and their prospects briefly discussed.

(1): So far, clinical experience with ATP treatment in cancer patients is limited. In 1995, following an uncontrolled phase I study in the US, we initiated a randomised clinical trial (RCT) with ATP infusions in cancer.1 A total of 58 patients with advanced non-small-cell lung cancer in the palliative treatment stage were randomised to ATP (continuous infusion for 30 h, max. rate 75 ug/kg.min, frequency 1×/2–4 wks) or control (palliative care only). Over a period of 6 months, ATP completely inhibited the loss in weight, fat mass, body cell mass, muscle strength and functional status (walking stairs, household activities, etc.) which were seen in the control group; also, ATP improved quality-of-life and fatigue.

The most intriguing finding was that ATP increased life expectancy from 3.5 to 9.2 months in a subgroup of cachectic lung cancer patients at a relatively early tumour stage (IIIB), Based on these results, further RCTs in cancer patients have been intiated. One such RCT (S. Beijer et al., presented at this meeting) is targeted at patients with mixed tumour types in the late palliative treatment stage cancer, and aims to improve weight, appetite, fatigue, physical disability and quality of life. ATP infusions in this protocol are applied in the home situation, which is feasible and safe provided a strict protocol is followed.

To (2/3) The strong favourable effect of ATP on life expectancy in lung cancer patients has triggered therapeutic studies, combining ATP with curative cancer treatment modalities such as radiotherapy and chemotherapy. At the same time, laboratory studies focus upon unravelling underlying mechanistic of the effects of ATP in cancer, especially the ATP-effects on therapy-induced inflammation and oxidative stress (Swennen et al., presented at this meeting).

We conclude that ATP holds promise for cancer treatment, not only for alleviating some of the devastating side effects of cancer, such as weight loss and fatigue, but also for increaing survival in selected patient groups. Moreover, ATP has potential to enhance the anti-tumour effects of current anti-tumour agents, while at the same time attenuating their side effects.

Design, synthesis and antitumor activity of 4′-Thionucleosides as highly potent and selective A3 adenosine receptor agonists

Lak Shin Jeong

Laboratory of Medicinal Chemistry, College of Pharmacy, Ewha Womans University, Seoul 120–750, Korea

A number of ligands have been synthesized and tested for binding affinity at the rat, sheep, and human A3 versus A1 and A2a receptors. Among these ligands, IB-MECA (1) was found to be a highly potent human A3 agonist (Ki = 1.8 nM), which is 900- and 1620-fold selective for human A3 versus either human A1 or A2a receptors, respectively. Introduction of chlorine at the 2-position of IB-MECA, resulting in the formation of Cl-IB-MECA (2) also increased binding affinity and selectivity. Compound 2 has been reported to display a Ki value of 1.4 nM and showed 900- and 3800-fold human A3 receptor selectivity versus human A1 and A2a receptors, respectively. Compounds 1 and 2 are in phase II clinical trials and preclinical trials as anticancer agents, respectively. Compound 2 is also used extensively as a pharmacological tool for studying A3 receptors. Open image in new window

On the basis of high binding affinity and selectivity of 1 and 2 on human A3 adenosine receptors, we designed and synthesized the 4′-thio analogues 3, since a sulfur atom may serve as a bioisostere of an oxygen atom. The synthesized 4′-thionucleoside derivatives exhibited dramatic increases in affinity and selectivity for human A3 receptors versus human A1 and A2a receptors. Among compounds tested, 2-chloro-N6-(3-iodobenzyl)-4′-thioadenosine-5′-methyluronamide (Thio-Cl-IB-MECA) inhibited the growth of human promyelocytic leukemia HL-60 cells by arresting cell cycle and induction of apoptosis. The growth inhibitory activity of thio-Cl-IB-MECA was also found to be related with the modulation of Wnt signaling pathway by measuring the levels of β-catenin, phosphorylated forms of GSK-β and Akt. Oral administration of Thio-Cl-IB-MECA exhibited highly potent antitumor activity in in vivo tumor xenograft model, indicating that it may be developed as a clinically useful anticancer agent.

Design, synthesis, and antitumor activity of the 4′-thionucleosides 3 as ultrapotent and selective agonists at the human A3 adenosine receptor will be presented in detail.

Growth regulation of tumor cells in hypoxia: Focus on A3 adenosine receptors

Stefania Merighi1, Annalisa Benini1, Prisco Mirandola2, Stefania Gessi1, Katia Varani1, Edward Leung3, Stephen Maclennan3, Pier Andrea BoreA1,4

1Department of Clinical and Experimental Medicine, Pharmacology Unit; University of Ferrara, 44100, Ferrara, Italy

2Department of Anatomy, Pharmacology and Forensic Medicine, Human Anatomy Section, University of Parma, 43100, Parma, Italy

3King Pharmaceuticals Research & Development, Cary, North Carolina 27513, USA

4Interdisciplinary Center for the Study of Inflammation; University of Ferrara, 44100, Ferrara, Italy. mhs@unife.it

Hypoxia appears to induce a program which shifts the cellular phenotype toward an increase in intracellular adenosine. Hypoxia-inducible factor-1 (HIF-1) is a key regulator of genes crucial to many aspects of cancer biology. Since the levels of both HIF-1 and adenosine are elevated within the hypoxic environment of solid tumors, we investigated whether adenosine may regulate HIF-1. In particular, we investigated the effect of A3 receptor antagonists on HIF-1, vascular endothelial growth factor and Angiopoietin-2 expression. Furthermore, this presentation will discuss new signaling pathways induced by A3 receptors in hypoxia and provide results of how human melanoma and glioblastoma tumor growth may be influenced through the adenosinergic system.

Inhibition of hepatocellular carcinoma growth by an A3 adenosine receptor agonist: De-regulation of PKB/Akt-NF-kB signaling pathway

Sara Bar-Yehuda, Avivit Ochaion, Shira Cohen, Faina Barer, Pnina Fishman Can-Fite BioPharma, Petach Tikva, Israel

The A3 adenosine receptor (A3AR) belongs to the family of the Gi-protein associated cell surface receptors and is highly expressed in various tumor cell types in comparison to the normal adjacent tissue. Lately it has been demonstrated that peripheral blood mononuclear cells (PBMNC) derived from colorectal patients also highly express A3AR in comparison to PBMNC derived from healthy volunteers. A3AR activation by the synthetic agonist suppress the development of melanoma, colon and prostate carcinoma in vitro and in vivo. The molecular mechanism is mediated via down-regulation of PKB/Akt which induces down-stream deregulation of the Wnt and the NF-kB signal transduction pathways.

In this study we followed A3AR expression in hepatocellular carcinoma (HCC) as well as in PBMNC both in human and in an animal model. The effect of an A3AR agonist administration on A3AR expression level, tumor development and signal transduction pathways involved were studied.

Utilizing paraffin embedded slides derived from human HCC samples, it was demonstrated that A3AR is highly expressed in HCC tumor tissues, in comparison to the normal adjacent tissue. The high expression of the A3AR in the tumor tissue was reflected in PBMNC derived from HCC patients. In addition, in tumor tissues as well as in PBMNC derived from N1S1 hepatoma bearing rats high A3AR expression level was noted in comparison to normal liver tissues. The expression level of the receptor was down-regulation upon oral treatment with the A3AR agonist Cl-IB-MECA (designated as CF102). These events were followed by a decreased levels of PKB/Akt, IKK, NF-κB and TNF-α, all members of the NF-κB signaling pathway. Moreover, GSK-3β was upregulated while β-catenin, Lef/Tcf and c-Myc expression level was down-regulated, demonstrating that deregulation of the Wnt signaling pathway took place. As a result, a marked inhibiting in the growth of the N1S1 tumors was observed in the CF102 treated animals.

Taken together we conclude that the A3AR may be developed as a new surrogate marker. In addition the receptor may be suggested as target to be activated by synthetic small and orally bioavailable agonists which inhibits HCC growth.

P2X7: A growth-promoting receptor

Francesco Di Virgilio

Department of Experimental and Diagnostic Medicine, Section of General Pathology, University of Ferrara fdv@unife.it

P2X7 is a receptor for extracellular nucleotides expressed by different normal cell types. Activation of P2X7 may result in stimulation of cell proliferation or induction of apoptosis, depending on the level of activation. We have previously correlated the level of P2X7 expression to disease severity in lymphocytes from patients affected by chronic lymphocytic leukemia (1). Furthermore, we have recently shown that expression of this receptor allows growth in the absence of serum (2). This growth-promoting effect appears to be mediated by three main factors: a) increased cytoplasmic Ca2+ levels; b) increased mitochondrial potential; c) enhanced ATP synthesis. P2X7-expressing cells also release larger amounts of ATP into the pericellular space. We have found that tumours, among which human neuroblastoma, express a higher level of P2X7 compared to non-tumour cells (3). Growth-promotion by P2X7 may be mediated via additional mechanisms involving release of growth stimulatory factors among which substance P and ATP itself.

This novel findings suggest that overstressing the well known cytotoxic activity of this receptor may be limitative and in certain conditions even misleading.

Regulation of astrocyte proliferation by P2Y and P2X purinergic receptors

{urJoseph T. Neary}, Yuan Kang, and You-Fang Shi

Research Service, Miami VA Healthcare System, Departments of Pathology, Biochemistry & Molecular Biology, and Neuroscience Program, University of Miami Miller School of Medicine, Miami, Florida 33125, USA jneary@med.miami.edu

Extracellular ATP enhances the mitogenic activity of fibroblast growth factor-2 (FGF2) in astrocytes1, but the molecular mechanism(s) and the P2 purinergic receptor type(s) involved are not well understood. As one approach to determine whether the potentiating effect of extracellular ATP involves cell cycle control mechanisms, we measured the expression of cyclins that are induced in different phases of the cell cycle in primary cultures of rat cortical astrocytes2. ATP potentiated the ability of FGF2 to stimulate expression of cyclin D1, a regulator of cell cycle entry, as well as cyclin A, a regulator of DNA replication. The potentiating effect of ATP on cyclin expression was attenuated by inhibiting activation of extracellular signal regulated protein kinase (ERK) which is coupled to FGF2 and P2 receptors. P2 agonist studies revealed that UTP enhanced FGF2-induced cyclin expression and mitogenesis whereas 2-methylthio ADP was ineffective, thereby suggesting a role for P2Y2 or P2Y4 receptors in the potentiation of astrocyte proliferation induced by FGF2. In contrast to the effect of UTP, we found that an agonist of P2X receptors, 2′,3′-O-(4-benzoyl)benzoyl-ATP (BzATP), inhibited FGF2-induced mitogenesis by 90%. This effect was not explained by a cytotoxic response to BzATP because at least 95% of astrocytes treated with BzATP remained viable. BzATP is an agonist of P2X7 as well as P2X1 and P2X3 receptors, but α,β-methylene ATP (in the media free of phenol red, an antagonist of P2X1 and P2X3 receptors3) did not inhibit FGF2-induced mitogenesis. In addition, dose-response studies demonstrated that 300 to 1000 micromolar ATP reduced the ability of FGF2 to stimulate DNA synthesis. These results indicate that P2X7 receptors mediate growth arrest in astrocytes. Consistent with opposing effects of P2Y and P2X receptors on mitogenesis, UTP stimulated a transient activation of ERK whereas BzATP stimulated a more sustained ERK signal. In addition, phospho-histone 3, a marker of mitosis during M phase of the cell cycle, was decreased by BzATP whereas UTP increased levels of phospho-histone 3 induced by FGF2. Collectively, these findings suggest that signaling by P2Y2/4 receptors enhance the ability of FGF2 to stimulate entry into a new cell cycle, DNA replication and mitosis by an ERK-dependent mechanism, whereas signaling by P2X7 receptors inhibits FGF2-induced mitogenesis in astrocytes and leads to growth arrest. Interactions between P2Y, P2X and polypeptide growth factor signaling pathways may have important implications for CNS development as well as injury and repair. [This work was supported by the National Institutes of Health (NS046651) and the Department of Veterans Affairs.]

Targeting the A3 adenosine receptor to combat cancer and autoimmune inflammatory diseases

Pnina Fishman, Avivit Ochaion, Shira Cohen, Faina Barer, and Sara Bar-Yehuda

Can-Fite BioPharma, Petach Tikva, Israel

CF101 is a small molecule adenosine derivative which binds specifically to the A3 adenosine receptor (A3AR). The latter is abundantly expressed on the surface of cancer and inflammatory cells whereas only low expression was found in normal cells. Activation of this receptor de-regulates key intracellular signaling pathways and inhibits the proliferation of cancer and inflammatory cells. We have built a drug development platform around this novel target and utilized synthetic A3AR agonists, with high affinity to the receptor, to treat cancer and inflammatory diseases.

CF101 inhibited the growth of melanoma, colon prostate and pancreatic carcinoma tumors in experimental mice models when given daily orally. The molecular mechanism involves de-regulation of the Wnt and the NF-kB signal transduction pathway, i.e., downregulation of PKA and PKB/Akt and up-regulation of GSK-3β which subsequently phosphorylates β-catenin leading to its ubiquitination. As a result, a decrease in the level of the 2 cell cycle progression genes, cyclin D1 and c-Myc takes place.

CF101 inhibits the clinical and pathological manifestations of adjuvant induced arthritis in rats and collagen induced arthritis in DBA mice. On the molecular level, CF101 deregulates the PI3K-PKB/Akt-NF-kB pathway thus leading to the inhibition of TNF-α and the induction of apoptosis of inflammatory cells. CF101 also showed efficacy as a disease modifier in additional experimental inflammatory models including colitis and multiple sclerosis.

CF101 is an orally administered molecule with excellent safety profile and specific activity against targeted pathological cells. This molecule is being tested in several clinical trials, including an early PhII study in patients with active rheumatoid arthritis. A recent interim analysis of this trial demonstrated provocative evidence of therapeutic activity and an excellent safety profile.

Pain and Nerve Transmission

Adenosine receptors in astrocytes, cytokines and neuroprotection

Knut Biber

Department of Medical Physiology, University Medical Center Groningen, University of Groningen, The Netherlands K.Biber@med.umcg.nl

High concentrations of adenosine are released in the brain under neuropathological conditions. Recent evidence suggest that stimulation of adenosine (A1, A2B or A3) receptors in glial cells induces the release of neuroprotective substances, indicating that also adenosine receptors in glia cells contribute to adenosine-dependent neuroprotection. The release of NGF in astrocytes for example is regulated by adenosine A1 receptor activation, whereas other adenosine receptors are involved in astrocytic release of neuroprotective cytokines. We have recently provided evidence that stimulation of adenosine A3 receptors in cultured astrocytes stimulates the release of the neuroprotective chemokine CCL2 (formerly known as MCP-1). This is based on pharmacological evidence since the selective adenosine A3 receptor agonist 2-chloro-N6-(3-iodobenzyl)-N-methyl-5′-carbamoyladenosine (CL-IB-MECA) induced the release of CCL2, whereas specific ligands for adenosine A1 or A2 receptors did not affect CCL2 release. Furthermore, CL-IB-MECA-induced CCL2 synthesis was inhibited by adenosine A3 receptor antagonists [4]. Earlier studies of our group have shown that stimulation of astrocytic adenosine A2B receptor is instrumental for the release of IL-6, a cytokine with well-known neuroprotective properties [1, 2]. Thus these data demonstrate that various adenosine receptor subtypes in astrocytes are responsible for the synthesis of different neuroprotective agents. Since these receptor subtypes display different affinities for adenosine it is suggested that depending on the concentration of extracellular adenosine specific patterns of neuroprotective substances are synthesized by astrocytes [3].

How cytokines like CCL2 or IL-6 protect neurons is not understood yet. We have recently obtained evidence that might explain the neuroprotective properties of IL-6. It is presented here that IL-6 enhances expression and function of neuronal adenosine A1 receptors in vitro and in vivo. It is therefore suggested that IL-6 increases neuronal survival in vitro and in vivo by enhancing the function of neuronal adenosine A1 receptors, the brains major protective system. The data presented here thus indicate a tight interplay between cytokines and the adenosinergic system in astrocytes and neurons with major impact in neuroprotection.

Adenosine regulation of neurotrophin synaptic actions

Ana M. Sebastião, M. José Diógenes, Paula I. Pousinha, Catarina C. Fernandes, António Pinto-Duarte, Catarina A. Gomes, Sandra H. Vaz and J. Alexandre Ribeiro

Institute of Pharmacology and Neurosciences, Faculty of Medicine and Institute of Molecular Medicine, University of Lisboa, Portugal

ATP and neurotrophins, in particular Brain Derived Neurotrophic factor (BDNF) have been implicated in generation and modulation of pain (e.g. Coull et al., 2005 — Nature 438: 1017–21). In contrast, adenosine has significant potential for alleviating various types of pain (see e.g. Hayashida et al., 2005 — J Anesth 19: 225–35). BDNF-induced long lasting changes in synaptic efficacy in the supraspinal descending pathway have been pointed out as one of the causes for BDNF-related development of persistent pain (Guo et al., 2006 — J Neurosci 26: 126–37). BDNF-induced changes in synaptic activity and efficacy also occur in other areas of the central nervous system, including hippocampus (e.g. Braham and Messaoudi, 2005 — Prog Neurobiol 76: 99–125). Synaptic actions of BDNF are facilitated by enhanced neuronal activity (see Nagappan and Lu, 2005 — Trends Neurosci 28: 464–71), depolarization (Boulanger and Poo, 1996 — Nature Neurosci 2: 346–51) and cyclic AMP (Boulanger and Poo, 1996 — Science 284: 1982–4). Depolarization induces the release of adenosine (e.g. Latini and Pedata, 2001 — J. Neurochem 79: 463–84) and enhanced neuronal activity facilitates adenosine A2A receptor mediated actions (Correia-de-Sá et al., 2000 — J. Neurochem 74: 2462–9). Since adenosine A2A receptors are coupled to cyclic AMP and since activation of these receptors can induce transactivation of BDNF TrkB receptors (Lee and Chao, 2001 — Proc Natl Acad Sci 98: 3555–60) we evaluated how activation of adenosine A2A receptors could influence synaptic actions of BDNF.

We observed that A2A receptor activation by released adenosine is an essential pre-requisite to observe an effect of BDNF upon excitatory synaptic transmission in the CA1 area of the hippocampus (Diógenes et al. 2004 — J Neurosci 24: 2905–13). The BDNF-induced enhancement of the release of acetylcholine at the hippocampus (Fernandes et al., Program No. 157.17. Abstract Viewer. SfN, 2005) and the BDNF-induced enhancement of neuromuscular transmission at the innervated phrenic-diaphragm (Pousinha et al., 2006 — Br. J. Pharmacol, proc sup, in the press) also require tonic activation of adenosine A2A receptors. Triggering BDNF actions by A2A receptor activation involves cyclic AMP and protein kinase A (Diógenes et al., 2004). BDNF inhibits GABA uptake from hippocampal nerve terminals but this action is independent of A2A receptor activation (Vaz et al., Br J. Pharmacol, Proc suppl, in the press). In the striatum, another neurotrophin, glial cell line-derived neurotrophic factor (GDNF) enhances the release of dopamine and this action is also modulated by A2A receptor activation (Gomes et al. 2005, J. Neurochem., 94, Supl 2, 264).

The above results clearly show that A2A receptors trigger synaptic actions of neurotrophins. Whether A2A receptors also facilitate BDNF-induced enhancement of pain mechanisms and whether A2A receptor mediated inhibition of adenosine A1 receptor functioning (Cunha et al., 1994 — Brain Res 649: 208–16) and of A1 tonic inhibitory actions (Pinto-Duarte et al., 2005 — J. Neurochem 93: 595–604) may influence adenosine actions on pain, deserves further investigation.

Work supported by Funda(cão para a Ciência e Tecnologia, Portugal. BDNF was a gift by Regeneron.

Expression, function and modulation of P2X3 receptors on trigeminal ganglion nociceptors

A. Nistri, E. Fabbretti, A. Fabbro, M. D'Arco, M. Simonetti and R. Giniatullin

Neurobiology Sector, International School for Advanced Studies (SISSA), Trieste, Italy nistri@sissa.it.

Painful stimuli from head tissues are transmitted to the brainstem via certain populations of trigeminal ganglion (TG) neuron. Such neurons can be activated by a wide range of signals which comprise locally released substances like ATP, or acid pH, mechanical stimulation etc. In chronic pain states including migraine, it is suggested that there is a process of sensitization of trigeminal neurons which become hyper-responsive to stimuli. In the attempt to understand the molecular mechanisms responsible for chronic pain sensitization, we developed a primary culture of rat or mouse TG neurons to test the role of various pain mediators over an extended time period. Cultures were prepared from TGs of 2 week old animals and grown for 1–4 days in vitro without adding exogenous NGF. The first issue was to characterize the neuronal population in culture and to compare it with the one normally found in ganglia in situ. The focus of this study was on ATP-sensitive P2X receptors and capsaicinsensitive TRPV1 receptors: both classes are important transducers of nociception on sensory neurons.

Rat or mouse β-tubulinIII positive neurons demonstrated good viability in vitro with <10% of them labeled against the early markers of apoptosis JNK or activated caspase3. Neurons were predominantly of small and medium diameter like those found in ganglia. Real time RT-PCR and Western blot analysis showed upregulation of P2X3 and TRPV1 receptors in culture versus ganglia, while P2X2 ones were unchanged. The number of neurons immunoreactive to P2X3 and TRPV1 receptors grew with more frequent co-localization. TRPV1 immunoreactivity was, however, comparatively low in mouse ganglia and cultures. Ca2+ imaging and whole-cell patch clamping showed functional P2X and TRPV1 receptors. The commonest effect was a fast desensitizing response mediated by P2X3 receptor activity and antagonized by the selective blocker A-317491. More rat than mouse neurons generated mixed responses indicative of more efficient heteromerization of P2X3 and P2X2 subunits. We next tested if algogenic substances could modify P2X3 or TRPV1 receptors in culture. Twentyfour h application of NGF (50 ng/ml) selectively upregulated P2X3 receptors without affecting TRPV1 receptors, while 10 µM serotonin application (24 h) largely potentiated TRPV1 function without changing P2X3 receptor activity.

These data show that TG cultures allow studying early adaptive changes of nociception-transducing receptors including their selective modulation by algogenic substances. This work is supported by grants from Telethon Foundation (GGP 04037) and MIUR (FIRB).

Functional consequences of P2X7 activation in astrocytes and microglial cells

F. Bianco, A. Colombo, R. Mele, M. Matteoli and C. Verderio

CNR-Institute of Neuroscience, University of Milano Department of Medical Pharmacology, Center of Excellence on Neurodegenerative Diseases, Milano, Italy c.verderio@in.cnr.it

Formation and shedding of vesicles from the plasmamembrane commonly occur in haematopoietic and immune cells and represents a process whereby signal molecules are released into the microenvironment. We have recently shown, in microglial cells, that ATP stimulation controls vesicle shedding through the activation of P2X7 receptors and that shed vesicles represent a pathway for the release of the pro-inflammatory cytokine IL1-beta. Isolation of shed vesicles, followed by IL-1 beta evaluation by a specific ELISA revealed the presence of the cytokine inside the vesicular organelles and its subsequent efflux into the extracellular medium. Western blot analysis revealed the presence of P2X7 receptors, pro- IL-1 beta and pro-caspase-1 in isolated vesicles, solubilized immediately after isolation. Processed bands of both IL-1 beta and caspase-1 were detected in vesicles exposed 30 min to ATP before lysis, suggesting that shed vesicles are the site of IL1-beta processing. ATP-binding cassette (ABC) transporters, a large family of proteins whose role is to translocate various substances across biological membranes, mediate the efflux of IL1-beta from shed vesicles, as cytokine release was suppressed by the general ABC transporter inhibitor glibenclamide. Since shedding also occurs in resting microglia not expressing IL1-beta, it may represent a cargo system for the release of other signalling molecules from these cells. For these reasons, we are carrying out a biochemical characterization of isolated vesicles, to further characterize the protein composition of shed vesicles and investigate the role and function of these organelles.

Intracellular signaling underlying ATP-induced chemotaxis of microglia

K. Ohsawa1, Y. Irino1, Y. Nakamura1, C. Akazawa1, K. Inoue2, and S. Kohsaka1

1Department of Neurochem., Natl. Inst. Neurosci., Tokyo, Japan

2Department of Pharmcol., Grad. Sch. Pharm. Sci., Kyushu Univ., Fukuoka, Japan kohsaka@ncnp.go.jp

Resident microglia exhibit ramified shapes in the normal brain, however, in response to pathological stimuli, they transform rapidly themselves into more motile amoeboid form called activated microglia and migrate toward the lesioned sites, where the accumulating microglia secret a variety of substances [1] to repair the tissues. Thus, microglial cell migration is an important initial step of amelioration of the damaged nervous system.

Extracellular ATP is known to regulate physiological functions of microglia [2]. Microglia possesses both ionotropic P2X receptors (P2X4R and P2X7R) and G-protein-coupled P2Y receptors (P2Y2R, P2Y6R and P2Y12R). We previously showed that ATP induced membrane ruffling and chemotaxis of microglia and suggested that the membrane ruffling is mediated by Gi/o-protein-coupled P2Y12R [3, 4]. We showed here that the ATP-induced chemotaxis of microglia is also regulated by the ionotropic receptor, P2X4R in addition to the P2Y12R.

Stimulation of G-protein-coupled receptor lead to activation of phospholipase C (PLC) and phosphoinositide 3-kinase (PI3K). We examined the effect of PLC and PI3K inhibitors on the formation of membrane ruffling and the chemotaxis of microglia following the stimulation by ATP. A PLC inhibitor inhibited both membrane ruffling and chemotaxis, while PI3K inhibitors suppressed only chemotaxis without inhibiting the membrane ruffling. These observations indicate that PLC activation is essential for both of the membrane ruffling and the chemotaxis, while the activation of PI3K is necessary only for the chemotaxis. Phosphorylation of Akt, which is known to be a downstream target of PI3K pathway, was enhanced by ATP stimulation. The increase in Akt phosphorylation was suppressed by chelating extracellular calcium. These results indicate that activation of PI3K pathway is modulated by the extracellular calcium influx suggesting a possibility that ionotropic P2XRs are involved in the PI3K activation. Therefore, we examined the effect of various P2XRs antagonists on the ATP-induced chemotaxis of microglia. TNP-ATP significantly inhibited the chemotaxis, but neither PPADS nor BBG were effective. Furthermore, we constructed the lentivirus vector expressing short hairpin RNAi against P2X4R and introduced the vector into microglia and showed that suppression of P2X4R reduced the ATP-induced chemotaxis of the cells. These results clearly indicate that P2X4R in addition to P2Y12R are involved in the ATP-induced chemotaxis of microglia.

P2X4-dependent neuropathic pain: A mechanism of the facilitation in pain transmission via microglia

Kazuhide Inoue and Makoto Tsuda

Department of Molecular and System Pharmacology, Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi, Fukuoka 812-8582, Japan inoue@phar.kyushu-u.ac.jp

There is abundant evidence that extracellular ATP and other nucleotides have an important role in pain signaling at both the periphery and in the CNS. The focus of attention now is on the possibility that endogenous ATP and its receptor system might be involved in neuropathic pain. Neuropathic pain is often a consequence of nerve injury through surgery, bone compression, diabetes or infection. This type of pain can be so severe that even light touching can be intensely painful; unfortunately, this state is generally resistant to currently available treatments. We recently reported that the expression of P2X4 receptors in the spinal cord is enhanced in spinal microglia after peripheral nerve injury, and blocking pharmacologically and suppressing molecularly P2X4 receptors produce a reduction of the neuropathic pain behaviour (Nature 2003; 424: 778–83). More recently, we have reported that brain-derived neurotrophic factor (BDNF) released from microglia by the stimulation of P2X4 causes the depolarizing shift in reversal potential of anion in LI neurons of rats with nerve injury (Nature 438, 1017–21, 2005), resulting in causing neuropathic pain. Understanding the key roles of these ATP receptors may lead to new strategies for the management of intractable chronic pain.

The Role of P2X7 and P2X4 in pain processing; common or divergent pathways?

I.P. Chessell1, J.P. Hatcher1, J.P. Hughes2, L Ulmann3, P. Green3, P.K. Mander2, A.J. Reeve1 and F.A. Rassendren3

1Pain Research, GlaxoSmithKline, Coldharbour Road, Harlow, Essex, UK

2Neuro-Cell Sciences, GlaxoSmithKline, Coldharbour Road, Harlow, Essex, UK

3Institut de Genomique Fonctionnelle, CNRS, Montpellier, France IainP.Chessell@gsk.com

The P2X7 receptor is unique in its ability to regulate the release of mature, biologically active interleukin-1β (IL-1β) [1]. We have demonstrated that in mice lacking this receptor, inflammatory and neuropathic hypersensitivity is completely absent to both mechanical and thermal stimuli, but that normal nociceptive responses are preserved [2].

The P2X4 receptor has also been implicated in pain processing; the expression of P2X4 is upregulated in the spinal cord in microglial cells following nerve injury [3]. Whilst inhibition of P2X4 receptor expression by antisense oligonucleotides has been shown to inhibit nerve-injury pain behaviour [3], the role of this receptor in inflammatory pain, and its role in regulating inflammatory mediators has not been explored.

In this study, we describe the phenotypic characterization of P2X4 knockout animals (−/−) in inflammatory pain, and preliminary characterization of inflammatory mediator regulation in these animals, and compare these with phenotypic and cytokine characterization in P2X7 −/− animals to aid elucidation of the role these P2X receptor subtypes play in pain.

A fully backcrossed colony of P2X4 −/− mice (n = 15 per group) was examined in the Freund's complete adjuvant (FCA) model of inflammatory hypersensitivity and compared with responses in P2X4 +/+ littermate controls (n = 15 per group). 24 hrs following intraplantar injection of FCA, significant mechanical hyperalgesia was observed in the P2X4 +/+ animals, but no significant hyperalgesia was observed in the P2X4 −/− cohort. These data are similar to those observed in the P2X7 −/− animals [2], so a preliminary analysis of cytokine expression profile was performed in the P2X4 −/− animals to provide initial insight into commonality of mechanisms between the two transgenic lines.

Following intraplantar FCA injection, analysis of paw samples from the P2X7 −/− animals revealed significant attenuation of FCA-induced increases in IL-1β and IL-10, but a significant increase in IL-6 (see [2]). Conversely, in the P2X4 −/− animals, there was no reduction in FCA-induced increases in IL-1β, and an increase in concentrations of IL-10.

Taken together, these data indicate the P2X4 −/− and P2X7 −/− animals share a common pain phenotype, but this phenotype appears to be conferred via different mechanisms. Further studies are required to confirm and elucidate the mechanistic differences, and also to explore cross-talk between P2X4 and P2X7 in both transgenic lines.

Round table on non-Adenine-Based Purines

Evidence for a patho-physiological and pharmacological role of guanine-based purines as a new extracellular signalling system

M.P. Rathbone1 and F. Caciagli2

1Department of Medicine, McMaster University, Hamilton, Ontario, Canada

2Department of Biomedical Sciences, “G. d'Annunzio” University of Chieti-Pescara, Chieti, Italy mrathbon@mcmaster.ca

In addition to the adenine-based purinergic intercellular signaling systems, involving adenine, adenosine and adenosine phosphates, over the last 15 years analogous guanine-based systems have been discovered. Most of these are involved in “trophic” effects, affecting the growth, differentiation and survival of various cells. Indeed, guanine-based purines act synergistically with certain growth factors such as NGF, and also stimulate the production and release from cells of several growth factors and cytokines. Guanine-based purinergic signaling has been particularly investigated in cells of the nervous system and muscle. However, in addition other tissues, including skin, respond to these compounds, indicating that, like adenine based purinergic signaling, guanine based signaling may be widespread throughout many cell types and organs.

Guanine-based purines are released from cells, and when cells are damaged the release increases substantially. Indeed, under conditions simulating ischemia, cells release more guanine-based purines than adenine based purines. Moreover, the extracellular concentration of the guanine-based purines is higher than that of the adenine-based counterparts.

There is evidence that in some cases guanosine produces its effects through entering cells and interacting with an NGF-inducible protein kinase. But there is also evidence that guanosine may interact with unique receptors on the surface of cells. Similarly, there is evidence that GTP may also have cell-surface receptors that mediate some of its effects. Moreover, guanosine is metabolized to guanine by the enzyme purine nucleoside phosphorylase (PNP). Experimental studies have indicated that exogenous extracellular guanosine is relatively persistent compared to adenosine, but a large proportion of guanosine is metabolized to guanine. Emerging evidence indicates that guanine may also have its own extracellular signaling system that is distinct from guanosine. Certainly, this would be of particular interest, since the enzyme that metabolizes guanine, guanine deaminase, shows 50 fold regional variations in brain. This degree of regional variation is characteristically associated with enzymes that degrade neurotransmitters.

It appears that the concept of intercellular signaling by guanine-based purines is now well substantiated. Since GTP, guanosine and guanine have different biological effects, different receptive mechanisms and likely different signal transduction mechanisms, it could be suggested the intriguing possibility that the extracellular interconversion of these guanine derivatives provides an extra layer of signal regulation by cells.

Guanosine stimulates remyelination in injured spinal cord through promotion of proliferation and differentiation of oligodendroglial precursor cells

S. Jiang1, M.P. Rathbone2, P. Giuliani3, F. Caciagli3 and P. Di Iorio3

1Department of Surgery, McMaster University, Hamilton, Ontario, Canada

2Department of Medicine, McMaster University, Hamilton, Ontario, Canada

3Department of Biomedical Sciences, “G. d'Annunzio” University of Chieti-Pescara, Chieti, Italy jiangs@mcmaster.ca

The commonest type of spinal cord injury is a crush, which causes a central area of damage within the spinal cord, disrupting long tract nerve fibers passing up and down the spinal cord. Commonly, in the penumbra around the injury some nerve fibers survive, but the oligodendroglia that myelinate them die. Therefore these fibers do not conduct impulses, or do not conduct them well. However, the spinal cord contains endogenous precursors of the myelin-forming oligodendroglia, which, under certain circumstances, can differentiate into mature oligodendroglia and myelinate axons. This raises the possibility that after injury there is insufficient signal to trigger the differentiation of the oligodendroglial precursor cells into mature oligodendroglia that are capable of remyelinating axons. Guanosine (GUO) stimulates proliferation and differentiation of many types of cells in vitro and exerts neuroprotective effects in the central nervous system. Thirty five days after a standard moderate crush injury rats have a chronic, stable, permanent neurological motor deficit. We treated these rats with intraperitoneal (i.p.) GUO, 8 mg/kg body weight/day for up to 14 days. Their motor function began to improve after three days and continued to improve throughout the duration of the treatment. The improvement in motor function correlated with the increase in myelination in the penumbra of the injury. The myelin was characteristic of central nervous system myelin. The myelin arose from NG2-positive oligodendroglial precursor cells that matured into myelin-forming oligodendroglia. These may contribute to the myelin-forming cells. We hypothesize that the injury may ‘prime’ the intrinsic adult pregenitor cells in the penumbra of the lesion to respond to guanosine. Since GUO was administered systemically we questioned whether it acted directly on the nervous system as GUO, or as a metabolic product, or whether it did not even enter the central nervous system but instead exerted its effects by stimulating production of some peripheral hormone or cytokine. We used capillary electrophoresis combined with the radioactivity measurement to evaluate the distribution and metabolism of GUO (8 mg/Kg body weight), containing an aliquot (about 0.005% of the total GUO) of [3H]-GUO (7.3 Ci/mmol) administered by i.p. injection. GUO entered lesioned areas of spinal cord to the same extent as the remainder of the central nervous system. Over-basal exogenous GUO (eGUO) rapidly increased in plasma, reaching 80% maximum after 7.5 minutes and peaked after 30 minutes, thereafter remaining constant for the next hour. The amount of eGUO and its metabolite guanine (GUA) increased in the central nervous system and adipose tissue. Fifteen minutes after injection the amount of cerebral GUO was nearly doubled in comparison with basal value and the ratio of eGUO:eGUA was about 2:1, but after 60 minutes the eGUO:eGUA ratio was <1. Meanwhile the total eGUO + eGUA continued to increase in brain, spinal cord and adipose tissue, but not in other tissues, where it remained stable. These data indicate that systemically-administered GUO may certainly stimulate oligodendroglial precursor cells. But as well, it raises the possibility that GUA may also play a role.

Metabolism of extracellular nucleotides and nucleosides in the vasculature

Gennady G. Yegutkin1, Tiina Henttinen2, Andrey Mikhailov2, Sergei Samburski1 and Sirpa Jalkanen1

1MediCity Laboratory and Department of Medical Microbiology, Turku University and National Public Health Institute, 20520 Turku, Finland

2Turku Centre for Biotechnology, University of Turku/Åbo Akademi University, POB 123, 20521 Turku, Finland gennady.yegutkin@utu.fi

Extracellular nucleotides and nucleosides trigger diverse immunomodulatory, prothrombotic and vasoactive signaling events in the vasculature. The duration and magnitude of purinergic signaling are thought to be mainly governed by ecto-nucleotidases expressed on endothelial and hematopoietic cells. In contrast to traditional paradigms that focus on cell-associated inactivating mechanisms, the existence of alternative pathways has received little attention. By using human vascular endothelial cells, normal and leukaemic lymphocytes, and cell-free serum as enzyme sources, we identified the following three groups of cell surface-associated and soluble activities: (1) nucleotide-inactivating enzymes, NTP diphosphohydrolase/CD39, nucleotide pyrophosphatase and ecto-5′-nucleotidase/CD73; (2) nucleotide-phosphorylating enzymes, adenylate kinase and nucleoside diphosphate kinase; (3) nucleoside-inactivating enzymes, adenosine deaminase and purine nucleoside phosphorylase. Evidence for this was obtained by using independent assays, including bioluminescent measurement of ATP metabolism, radio-TLC and reverse-phase HPLC analyses. The comparative measurements of enzymatic activities indicated the predominance of the nucleotide-inactivating pathway on the endothelial cells, whereas the lymphocytes are generally characterised by counteracting ATP-regenerating/nucleoside-eliminating phenotype. Next, given that ecto-adenylate kinase represents the only AMP-converting pathway on ecto-5-nucleotidase negative leukaemic T- and B-lymphocytes, we employed this ectoenzyme as an intrinsic probe for accurate “sensing” ATP in the pericellular space. This novel enzyme-coupled approach, together with confocal imaging of putative ATP stores, suggests that, along with predominant ATP accumulation within cytoplasmic granules, micromolar concentrations of ATP are constitutively retained on lymphoid cell surface without significant convection into bulk milieu. In conclusion, identification of the extensive network of enzymes directionally regulating active cycling between ATP and other nucleotides and nucleosides provides a novel insight into the purinergic homeostasis in the vasculature. Furthermore, the ability of lymphocytes to maintain micromolar ATP halo may represent an important route initiating signaling cascades within immunological synapses and facilitating leukocyte trafficking between the blood and tissues.

The guanine-based purinergic system as a new target for neuroprotection against glutamatergic excitotoxicity

Diogo O. Souza

Department of Biochemistry ICBS, Institute of Basic Sciences of Health, “Rio Grande do Sul” Federal University, Porto Alegre, RS, Brazil diogo@ufrgs.br

Glutamate is the main excitatory neurotransmitter in mammalian CNS, essential for brain activities, as those involved in development, aging, memory, and adaptation to the environment. However, hyper-activation of the glutamatergic system may be potentially neurotoxic, involved in the pathogenesis of various acute and chronic brain injuries.

Our group has given strong evidence that the guanine-based purinergic system is effectively neuroprotective against glutamate toxicity, in acute and chronic animal models, both in vitro and in vivo studies. Although the administration of guanine derivatives (GD) exerts neuroprotection, our results strongly indicate that the active compound is the nucleoside guanosine (GUO).

In in vivo studies carried out in rat and mouse, GD protect against brain damage caused by hyper-activation of the glutamatergic system. Indeed: i) chronically, GMP administration in rat striatum protects cells against death caused by quinolinic acid (QA); ii) acutely, GMP or GUO i.c.v., i.p. or orally administered protect against seizures induced by QA (or α-dendrotoxin). In in vitro studies carried out in brain slices, GUO protects cell against death caused by in vitro ischemia.

Searching for mechanisms implicated in this neuroprotection, we demonstrated that: i) GUO stimulates the astrocytic glutamate uptake (in astrocyte cultures and brain slices), the main process involved in endogenous neuronal protection; ii) QA induced-seizures decrease glutamate uptake by cortical brain slices and this decrease is reversed by GUO when it acts as anticonvulsant; iii) Brain ischemia decreases glutamate uptake by hippocampal slices and i.p. GUO administration prevents this decrease. Thus we propose that the stimulatory effect on glutamate uptake is involved in the neuroprotective actions of GUO.

These results encourage further studies aiming at the therapeutic use in humans of GUO in acute (hypoxia, ischemia, brain traumatism) and chronic (neurodegenerative diseases) brain injuries involving glutamate excitotoxicity.

The role of uridine kinase and uridine phosphorylase in the cross regulation of uracil and adenine nucleotides salvage synthesis in rat and human brain

Piero L. Ipata and Maria G. Tozzi

Department of Biology, University of Pisa, Via San Zeno 51, 56127 Pisa, Italy ipata@dfb.unipi.it

A wealth of experimental data support the idea that uridine and its nucleotides (NT) have critical functions that help regulating a variety of biological systems [1, 2]. In cultured cells the concentration of uridine (≈3–8 µM) is sufficient to meet the pyrimidine requirement. At 12 µM the UTP pool doubles, and the de novo pyrimidine synthesis is almost 100% inhibited, pointing for the importance of the nucleoside as a precursor in the pyrimidine salvage pathway. In contrast, purines are salvaged from their bases. Both in bacteria and in mammals the pathways for de novo pyrimidine and purine synthesis from simple precursors are cross regulated, i.e. accumulation of a purine NT (ATP) activates pyrimidine synthesis and inhibits that of purines, and vice versa. However, it is well established that several tissues and organs, including brain, rely more heavily in the salvage synthesis of NT from preformed purine and pyrimidine rings, rather than on de novo synthesis. This raises the following question: how do these districts maintain the right balance between their purine and pyrimidine pools? Recent in vitro evidence suggests that in rat and human brain cytosol the two processes of purine and pyrimidine NT salvage respond in an opposite manner to the fluctuation of intracellular [purine nucleoside triphosphates]/[pyrimidine nucleoside triphosphates] ratio (PUR/PYR ratio). The metabolic sensor is a two enzyme system, composed of uridine kinase (Uk) and uridine phosphorylase (UPase), which contributes in maintaining uridine homeostasis. When the activity of Uk becomes inhibited by relatively high UTP and CTP concentrations (low PUR/PYR ratio), uridine accumulates and the equilibrium of the UPase reaction is shifted towards uridine phosphorolysis [3]. The ribose1-P produced forms PRPP [4], an obligate substrate of purine salvage. At a high ratio these effects are reversed. Uk is fully active, and pyrimidine salvage is favoured over purine salvage. Uridine might therefore be considered as a signal of pyrimidine abundance and a ribose carrier for purine nucleotides salvage.

An additional factor which might contribute in balancing the purine and pyrimidine salvage processes is the “oxypurine cycle,” in which hypoxanthine (or guanine) is converted to IMP (or GMP) by HPRT, at the expense of PRPP. IMP (or GMP) is then dephosphorylated to inosine (or guanosine) by cN-II, and inosine (or guanosine) is converted back to hypoxanthine (or guanine) and Rib 1-P by PNP [5]. We notice that both PRPP, the “fuel” of the cycle, and Rib 1-P, its main product, are essential in NT salvage synthesis. PRPP is an obligate precursor of purine salvage, while the only known pathway for salvaging uracil is its Rib 1-P mediated ribosylation to give uridine, followed by multiple phosphorylation steps. The observation that in vitro uracil is readily converted into uridine in the presence of PRPP, provided hypoxanthine is present in the reaction mixture, led to the proposal that Rib 1-P needed to ribosylate uracil might well arise from PRPP breakdown occurring in the oxypurine cycle [5].

Towards a patho-physiological role of non adenine purine nucleobases in the central nervous system

F. Caciagli1, P. Ballerini1, R. Ciccarelli1, P. Di Iorio1, A. Poli2, F. Licastro2, S. Jiang3 and M.P. Rathbone4

1Department of Biomedical Sciences, “G. d'Annunzio” University of Chieti-Pescara, Chieti, Italy 2Department of Experimental Pathology and Biology, University of Bologna, Bologna, Italy 3Department of Surgery, McMaster University, Hamilton, Ontario, Canada and 4Department of Medicine, McMaster University, Hamilton, Ontario, Canada. f.caciagli@dsb.unich.it

A growing evidence indicates that guanosine (GUO) exerts multiple neuroprotective activities. Indeed, it controls glutamate-induced excitotoxicity; causes astrogliosis and activates oligodendroglial cell precursors; modulates expression and function of some glial K+, Cl and water channels; stimulates ApoE production and cholesterol efflux from astrocytes; enhances trophin production and exerts anti-apoptotic activity. These effects are not substantially affected by cell pre-treatment with nucleoside and nucleobase transporter inhibitors, thus indicating that they are due to the extracellular activity of GUO, likely mediated by specific plasma membrane receptor(s).

To correctly evaluate the effects caused by the extracellular GUO (eGUO) and to individuate receptor(s) mediating these effects, it is crucial to know whether and how eGUO is metabolized and affects the breakdown and the effects of the extracellular adenine-based purines.

We found that eGUO stimulates the efflux of purine and pyrimidine nucleotides from cultured glial cells and it is well known that the purine/pyrimidine nucleotide ratio as well as the concentrations of ATP and Pi control the activity of nucleotidases which play a pivotal role in the purine metabolism, especially in the nucleoside formation. On the other hand, the receptor-mediated activity of released nucleotides as well as that of the de novo formed nucleosides (especially adenosine) affect, in synergistic or, sometimes, in antagonistic manner, the effects caused by eGUO. Indeed, eGUO up-regulates the expression of P2Y2 receptors thus amplifying the ATP and UTP effects mediated by these receptors and extracellular adenosine, like eGUO, contributes, via A1 receptors, to the production of trophic factors whereas it affects, in opposite manner, the apoptotic death of glial cells via A1 and A3 receptors, respectively.

Moreover, we individuated the presence of a purine nucleoside phosphorylase (PNP) on the external surface of the plasma membranes of glial cells (ecto-PNP) which, in nature and in presence of Pi, catalyzes the phosphorolysis of non adenine-based nucleosides (GUO and Inosine) thus producing the corresponding nucleobases (guanine and hypoxanthine). In this context, we recently addressed our attention to establish whether the ecto-PNP-mediated formation of purine nucleobases could also interfere with the effects caused by the respective nucleosides, by especially evaluating (in cells provided or not provided with PNP) the possible effects caused by extracellular guanine (eGUA) and by comparing them with those caused by eGUO, in the same cells. We found that eGUO and eGUA, by different molecular mechanisms, affect cell proliferation/death/apoptosis and learning/memory processes. In particular, eGUA (8 mg/kg orally administered) almost completely counteracted the L-NAME-induced loss of memory in a passive avoidance task carried out in rats. Accordingly, an exonic polymorphism in PNP gene (transition G/A at +3052 position), compatible with changes in GUA formation, resulted to be coupled with a rapid and severe decay of cognitive activity (MMSE test) in patients with clinical diagnosis of Alzheimer's disease.

Ectonucleotidases Structure and Function

Contibution of NPP-type ectophosphodiesterases to extracellular nucleotide metabolism

Mathieu Bollen

Division of Biochemistry, Faculty of Medicine, Catholic University of Leuven, B-3000 Leuven, Belgium

The extracellular level of nucleotides is dynamically controlled by the action of ecto-nucleoside diphosphokinases, ectoadenylate kinases and ectonucleotidases [1, 2]. There exist four structurally unrelated families of ectonucleotidases, namely E-NTPDases, alkaline phosphatases, NPP-type ectophosphodiesterases and 5-nucleotidase. The family of NPPs belongs to the superfamily of phospho-/sulfo-coordinating metalloenzymes that all have a similar catalytic fold and action mechanism. The NPP-family consists of seven members but only three of these, NPP1-3, are known to hydrolyse nucleotides. They are expressed as secreted (NPP2) or transmembrane proteins (NPP1, NPP3). In addition to the catalytic domain, they contain one nuclease-like domain and two somatomedin-B like domains, which are essential for the expression of catalytic activity.

The enzymatic action of NPP1-3 (in)directly results in the termination of nucleotide signaling, the salvage of nucleotides and/or the generation of new messengers like ADP, adenosine or pyrophosphate [2]. NPP1 and NPP3 are only known to hydrolyze nucleotides but NPP2 also acts as an extracellylar lysophospholipase-D. The simultaneous hydrolysis of lysophospholipids and nucleotides generates products that have the potential to synergistically promote cell motility.

Progress on the structures and functions of nucleoside triphosphate diphosphohydrolases and a calcium activated nucleotidase

Terence L. Kirley

Department of Pharmacology and Cell Biophysics, University of Cincinnati, P.O. Box 670575, Cincinnati, OH 45267-0575, USA terry.kirkley@uc.edu

My laboratory has studied aspects of several ecto-nucleotidases for more than a decade. Recently, we have focused on three such proteins—two members of the human Nucleoside TriPhosphate Diphosphohydrolase (NTPDase) family of nucleotidases, and one human Calcium Activated Nucleotidase (CAN) with homology to the soluble nucleotidases used by insects to enable blood feeding by hydrolyzing host ADP.

The two NTPDases are NTPDase3, a cell surface, integral membrane protein, whose likely function is modulation of purinergic signaling, and NTPDase6, which exists as a membrane-bound form in internal membranes, but can also be secreted as a soluble form by cleavage of its N-terminal signal peptide. Studies on the soluble NTPDase6 have included methods of expression and refolding, as well as disulfide bond determination, and attempts to develop specific peptide inhibitors for the NTPDases using phage-display technology. Studies on NTPDase3 have centered on mutagenic approaches to delineate residues and regions important for nucleotidase activities and for oligomer formation and stability. We developed computational models for the 3-D structure of the extracellular portions of the NTPDases, and have re-interpreted previous site-directed mutagenesis results using these models. We also propose a speculative cartoon model for how movements in the transmembrane domains of cell-surface NTPDases (demonstrated by Dr. Guidotti_s group) might be linked to movements in the active site lobes of the 3-D model, mediated via several conserved proline residues near the juncture of the extracellular and transmembrane domains. The extracellular active site lobes predicted by the 3-D models are postulated to undergo hinge-like movements based on the homology of NTPDases with the actin superfamily of proteins, which exhibit butterfly-like domain movements.

Brain is one of the few tissues where NTPDase3 is relatively abundantly expressed. Therefore, we immunolocalized NTPDase3 in adult rat brain, and found it to be associated with neurons and mostly limited to midline regions of the forebrain and midbrain. Based on co-localization with hypocretin-1/orexin-A cells and fibers in the hypothalamus, we hypothesize that NTPDase3 may modulate feeding and sleep-wake behaviors under purinergic control in this brain region.

A crystal structure of soluble human Calcium Activated Nucleotidase (CAN) has been published. However, we are investigating the possibility of multiple structural and functional roles for Ca2+ in this calcium-binding nucleotidase, as well as studying the newly discovered Ca2+-dependent dimerization that occurs in this human CAN.

This work was supported by NIH grants HL59915 and HL72882. The transmembrane domains of E-NTPDase 1, CD39, regulate its enzymatic activity

Alison Grinthal, Sari Paavilainen, Scott Jones, James Wu and Guido Guidotti

Department of Molecular and Cellular Biology, Harvard University, 7 Divinity Avenue, Cambridge, Massachusetts, USA guidotti@fas.harvard.edu

CD39 is anchored to the plasma membrane by two transmembrane domains, one at each end of the protein. Removal of either or both transmembrane domains or disruption of their native state by detergent solubilization reduces activity by 90%, indicating that native function requires both transmembrane domains to be present and in the membrane. The transmembrane domains are also involved in dimerization of CD39. Here we report the results of experiments on the oligomeric state of CD39 in the plasma membrane of COS7 cells. We also investigate the interactions between the transmembrane domains of CD39 by mutation of selected residues analyzed by yeast selection, by disulfide crosslinking of cysteines introduced within the transmembrane helices and by manipulation of the lipid bilayer. The results indicate that the principal interactions between the transmembrane domains are intramolecular, that the helices are highly mobile, and that these dynamic motions are required for enzymatic activity and regulated by the status of the lipid bilayer.

Ecto-ATPases and Metabolism

Co-ordinated regulation of P2-receptor signaling by membrane bound NTPDases

J. Sévigny, F. Ben Yebdri, M. Fausther, G. Kauffenstein, F. Kukulski, E.G. Lavoie, J. Lecka, S.A. Lé vesque, M.N. Munkonda and J. Pelletier

Centre de recherche en Rhumatologie et Immunologie, Université Laval, Ste-Foy, Québec, Canada Jean.Sevigny@crchul.ulaval.ca

Through the activation of different and specific P2-receptors, extracellular nucleotides (ATP, ADP, UTP, UDP) regulate a variety of biological functions. The concentrations of these molecules are tightly regulated at the cell surface by ectonucleotidases. The most important family of enzymes that dephosphorylate ATP and other extracellular nucleotides are the members 1, 2, 3 and 8 of the ecto-nucleoside triphosphate diphosphohydrolase (E-NTPDase) family. These enzymes sequentially hydrolyze the terminal phosphate residue of triphospho- and diphosphonucleosides, and, in conjunction with ecto-5′-nucleotidase, facilitates the formation of adenosine, another important biological messenger. NTPDases 1,2,3 and 8 differ in their catalytic abilities to dephosphorylate nucleotides, as for example ATP. Using this substrate, NTPDase1 gives a rapid accumulation of AMP with little formation of ADP. In contrast, NTPDase2 hydrolyzes poorly the diphosphonucleoside derivative and accumulates it in its immediate environment. To give further options to the cells to tightly regulate nucleotide levels, NTPDases 3 and 8 generate a transient accumulation of ADP (a potent ecto-5′-nucleotidase inhibitor) with a late formation of AMP (the substrate of ecto-5′-nucleotidase).

These enzymatic properties are certainly of importance in the definition of functions of NTPDases. Another important parameter in the identification of functions is the cellular localization. With a set of antibodies that we have developed, we observed that NTPDase1 is mainly located on vascular endothelium, smooth muscles, as well as on macrophages. A number of experiments clearly demonstrated the function of NTPDase1 in platelet aggregation and thromboregulation. With mice deficient in NTPDase1 expression, preliminary data suggest a function of the enzyme in vascular smooth muscle cell contraction. Other recent data suggest that monocytic NTPDase1 may be involved in inflammation and cytokine secretion such as IL-8. NTPDase2 is localized to the adventitia of large blood vessels and on the external surface of some capillaries, in nerve structures, and in portal fibroblasts where a function in the control of cell proliferation has been proposed. NTPDase3 have been recently localized to neurons by Dr. T.L. Kirley and co-workers. We have also observed high expression of the enzyme in few epithelia of the digestive system as well as in Langerhans islet cells of the pancreas. Finally, we have demonstrated that NTPDase8 is the long known liver canalicular ecto-ATPase.

Localization of these ectonucleotidases in different cells and tissue compartments suggests distinct roles of these enzymes. In this presentation, I will discuss about the characterization and function of plasma membrane bound NTPDases, namely NTPDase1,2,3 and 8.

Ecto-5′-nucleotidase (CD73) and mucosal inflammation

Sean P. Colgan1, Linda F. Thompson2, Andreas Robinson1 and Nancy A. Louis1,3

1Center for Experimental Therapeutics, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts USA 2Immunobiology and Cancer Research Program Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma, USA 3Division of Newborn Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts USA colgan@zeus.bwh.harvard.edu

Sites of inflammation are characterized by significant shifts in metabolic activity1. Shifts in energy supply and demand can result in diminished delivery and/or availability of oxygen, leading to inflammation-associated tissue hypoxia and metabolic acidosis. These shifts in tissue metabolism, as indicated by previous studies, are frequently associated with vasculitis and profound recruitment of inflammatory cell types, particularly myeloid cells such as neutrophils (PMN) and monocytes. Under such conditions, we and others have observed activation of the global hypoxia regulatory transcription factor hypoxia-inducible factor (HIF). A significant HIF target gene during mucosal inflammatory disease (e.g. inflammatory bowel disease, colitis) is CD732. Previous studies have determined that CD73 is regulated by HIF at the gene promoter level3. Since tissue profiling studies revealed highest expression of CD73 in mucosal tissues, particularly the intestine4, we have addressed the role of CD73 in model inflammatory disease in vitro and in vivo. Studies using intestinal epithelial cells have revealed that high expression of CD73 is protective for a number of epithelial properties (e.g. barrier function, ion transport), likely through generation of adenosine at the epithelial apical surface. Inflammatory studies in mice conditionally lacking the alpha subunit of HIF-1 in intestinal epithelial cells have shown that the loss of functional HIF-1 correlates with decreased expression of CD73 and enhanced inflammatory disease in these mice. More recent studies have addressed the role of CD73 colitic disease progression. These ongoing studies have revealed that the induction of colitis increases epithelial expression of CD73 mRNA and activity during the acute phase colitic disease. Studies utilizing Cd73-null mice4 have revealed that relative that the loss of CD73 correlated with more severe clinical symptoms of colitis (weight loss, colon length), and increased levels of inflammatory markers (neutrophil numbers). Mechanisms of increased susceptibility with decreased CD73 will be discussed. Taken together, these studies provide unique insight into tissue microenvironmental changes during model inflammatory disease and identify CD73 as a critical control point during mucosal insult.

Structure and function of ecto-5′-nucleotidase based on structural studies of the bacterial 5′-nucleotidases

Norbert Sträter

Center for Biotechnology and Biomedicine, Faculty of Chemistry and Mineralogy, University of Leipzig, Deutscher Platz 5, 04103 Leipzig strater@bbz.uni-leipzig.de

On the basis of primary sequence alignments it is clear that the bacterial 5′-nucleotidases (5′-NT) are related to the vertebrate ecto-5′-nucleotidases. The bacterial enzymes are present as monomers, whereas the ecto-enzymes form homodimers. A crystal structure is available for E. coli 5′-NT [1]. The enzyme consists of two domains: The N-terminal domain (residues 25–342) binds two metal ions and contains an Asp-His dyad, which are important for the catalytic activity. This domain is related to other known enzyme structures of the calcineurin superfamily of dimetal phosphoesterases. The C-terminal domain (residues 362–550) has a unique structure, which has so far not been found in other protein structures. This domain provides the binding site for the adenosine moiety of the substrate. Thus, the active site is located between the two domains.

In the crystal structures, the protein has been characterised in two conformations, which differ in the relative orientation of the two domains [2]. The domain movement can be described as a rotation of the C-terminal domain around an axis, which passes through the center of the C-terminal domain. A rotation of up to 96° is necessary for the change between the inactive open and the active closed conformation. The domain rotation is necessary for the catalytic action of the enzyme, presumably to allow for substrate binding and product release [3, 4].

A sequence alignment of E. coli 5′-NT and the mammalian ecto-enzymes shows that both domains are conserved and a homology model can be built for ecto-5′-NT [5]. The ligands for the two catalytic metal ions are all conserved in ecto-50-NT with the exception of M1-ligand Gln254 which corresponds to Asn245 in ecto-5′-NT. It appears unlikely that the shorter side chain supports metal coordination of Asn245, but the Asn245 carboxamide group may bind a water molecule which is coordinated to M1. Structures of E. coli 5′-NT in complex with the substrate analogue inhibitor α,β-methylene-ADP revealed the characteristics of the substrate binding pocket of the C-terminal domain. A central feature is the hydrophobic stacking interaction of the adenine moiety, which is sandwiched between two phenylalanine side chains. The N1 nitrogen of the adenine ring is recognized by the carboxamide group of Asn431. The ribose group of the nucleotide is bound by Asp504, the backbone oxygen of Gly458 and Arg410.

Of these residues, the two phenylalanines are conserved in the ecto-enzymes or in electric ray 5′-NT are replaced by a tyrosine, which should also support the stacking interaction. Asn431 is replaced by a glycine residue. There is no other side chain nearby in the model which could replace the asparagine. Most likely, a water molecule interacts with N1 of the adenine ring. The aspartate residue bound to the hydroxyl groups of the ribose is conserved, as well as two of the three arginines. Arg375 of the E. coli enzyme is replaced by a serine. Thus, from the differences in the substrate binding site it is not obvious how the stronger substrate specificity of the mammalian enzymes towards AMP (vs. ADP and ATP) is achieved.

The key regulators of skeletal mineralization: TNAP, NPP1 and ANK as therapeutic targets for the treatment of osteoarthritis, hypophosphatasia and arterial calcification

José Luis Millán

Burnham Institute for Medical Research, La Jolla, California, 92037 USA millan@burnham.org

The mechanisms that regulate tissue calcification are of fundamental importance, as they ensure that ossification proceeds normally in the skeleton while pathological or ectopic calcification is prevented elsewhere in the body. Alterations in these regulatory mechanisms, either due to genetic defects or as a result of aging, lead to disease, such as the development of vascular calcification and osteoarthritis. Vascular calcification is a common occurrence in aging, atherosclerosis, diabetes, renal failure, aortic stenosis and prosthetic valve replacement. Osteoarthritis refers collectively to a number of disease states that affect the properties and functionality of the joints. Our work has shown that mice deficient in the nucleoside triphosphate pyrophosphohydrolase-1 (NPP1) isozyme (Enpp1−/− mice) or in the Ankylosis protein (ank/ank mutant mice) both have a deficit in inorganic pyrophosphate (PPi) production. PPi is a potent mineralization inhibitor and both these mutant mice share phenotypic features of softtissue ossification, corrected by simultaneous deletion of the tissue-nonspecific alkaline phosphatase (TNAP) gene (Akp2−/−)1. Simultaneous inactivation of the Akp2 gene in Enpp1−/− and ank/ank mutant mice led to normalization of inorganic pyrophosphate (PPi) levels, which in turn led to the normalization of depressed osteopontin (OPN) levels and importantly, correction of their calcification abnormalities [2, 3]. Enpp1−/− and ank/ank mice also display arterial calcification similar to that seen in human patients with mutations in the NPP1 gene. Mice deficient in matrix Gla protein (Mgp−/−) mice are also characterized by ectopic calcification of all arteries in addition to bone abnormalities. When crossed to the Hyp “hypophosphatemic” mice, Mgp−/− mice never developed arterial calcification and had a normal life span [4]. Similarly, [ank/ank; Hyp] double mutant mice had a normal life span, as opposed to ank/ank mice that usually die around 6 months of age. Also, injection of PPi into Mgp−/t mice corrected their arterial calcification and normalized their life span. These results indicate that lowering the extracellular Pi/PPi ratio is sufficient to prevent the appearance of an osteoarthritis-like phenotype and of the arterial calcification in these different mouse models. Recent characterization of Opn−/− and [Akp2−/−; Opn−/−] mice has suggested that OPN may in fact be a more important mineralization inhibitor than PPi, and that the effects of modulating PPi levels are carried out by the changes in OPN levels [5]. Our work currently focuses on developing therapeutic drugs to target the function of TNAP and NPP1 as a strategy to normalize the extracellular Pi/PPi levels and prevent/correct ectopic calcification. Funded by grants DE 12880 and AR 47908 from the National Institutes of Health, USA.

Adenosine and Neuroprotection

A2a adenosine receptors regulate CNS responses to ethanol and addicting substances: recent advances from cell biology to behavior

Ivan Diamond

Vice_President, Neuroscience CV Therapeutics Palo Alto, California, USA

Very high levels of Gs(olf)-coupled A2a adenosine receptors (A2a) are found in the nucleus accumbens (NAcb), a mesolimbic brain reward region involved in alcoholism and addiction. Ethanol inhibits adenosine re-uptake and increases extracellular adenosine levels, thereby activating A2a receptors. Activation of A2a receptors promotes stimulation of adenylyl cyclase (AC) and cAMP/PKA signaling. In the NAcb, A2a receptor expression occurs postsynaptically on the same neurons expressing Gi-coupled D2 dopamine (D2), mu opioid (MOR) and cannabinoid (CB1) receptors. All of these receptors are involved in addiction. Activation Gi-coupled receptors for 30 min. ordinarily inhibits cAMP production. By contrast, we show that a 10 min. activation of Gi-receptors involved in addictive behaviors acts synergistically with A2a adenosine receptors to stimulate cAMP/PKA signaling and induction of CRE-mediated gene expression. In all instances, synergy is mediated by free Gi_betagamma subunit stimulation of AC II and IV. Synergistic stimulation of cAMP production requires A2a adenosine receptors. Importantly, knockdown or inhibition of the betagamma/cAMP signaling pathway prevents synergy in primary NAcb neurons and eliminates addictive behaviors. Relapse is the most serious complication of addiction that prevents effective medical treatment of human addicts. In rats, A2a receptor antagonists administered directly into the NAcb or indirectly by i.p. injection reduces voluntary ethanol drinking and eliminates reinstatement of heroin-seeking behavior in rats. These findings in a rodent model of human relapse suggest that A2a adenosine receptor antagonists may be useful therapeutic agents in the treatment of alcoholism and addiction. (Supported by grants from NIAAA and the Department of the Army).

Adenosine A2A receptors and Parkinson's disease — Effects of Istradefylline on functional models of Parkinson's disease

Tomoyuki Kanda, PhD.

Department of CNS Research, Pharmaceutical Research Center, Kyowa Hakko Kogyo Co., Ltd.

Currently, dopamine replacement therapy is most effective treatment for the motor symptoms of Parkinson's disease(PD) [1]. However, its association with the development of motor complications such as dyskinesia and wearing-off phenomenon, limits its usefulness in late stages of the PD [2]. Adenosine A2A receptors are abundantly localized to the indirect striato-pallidal output pathway, which is one of two major striatal output pathwayscontroling motor behavior via basalganglia network. Istradefylline is a novel selective adenosine A2A receptor antagonist currently in Phase III clinical trials in PD. The results from Phase II clinical trials has demonstrated that the coumpaund provided a clinically meaningful reduction in OFF time and an increased ON time with non-troublesome dyskinesia in the L-dopa treated patients with motor complications [3].

In PD, loss of dopaminergic neurons in substantia nigra causes imbalance in neuronal activity between the indirect strato-pallidal and direct striato-nigral pathways by loss of dopaminergic influences onto striatal medium spiny neurons gerenrating each pathway. This imbalance induces entire abnormality of basal ganglia network, resulting in abnormal movements (i.e., hypokinetic motor control). It has been demonstrated that the adenosine A2A receptor activation causesinduces excessive activation of the indirect striato-pallidal pathway via the dual modulation in the striatum and globus palidus [4]. Therefore, blockade of the A2A receptors should decrease such an excessive excitability of the indirect pathway and alleviated the motor symptoms of PD. The results from studies in animal models of PD have demonstrated that istradefylline increases locomotor activity and decreases motor disabilities without inducing involuntary movements in animals rendered dyskinetic by previous exposure to L-dopa [5]. In MPTP treated non-human primate(marmoset) models of PD, istradefylline was administered with a low dose of L-dopa for 21 days to animals primed to exhibit dyskinesia. The amplitude of involuntary movements observed was not greater than that produced by L-dopa alone in MPTP treated marmosets. In addtion, istradefylline shows significant reduction of L-dopa induced dyskinesias compared with L-dopa alone [6]. Results of studies on animal models of PD indicate that istradefylline did not exacerbate L-dopa-induced dyskinesia, and suggested that chronic coadministration of istradefylline with L-dopa may decrease dyskinesias caused by L-dopa. The results from functional models, providing relevance to clinical study outcomes in PD patients, described that adenosine A2A antagonists appear to be most promising as the first major nondopaminergic therapy for PD.

Adenosine A2A Receptors Modulate Psychomotor Activity and Brain Injury by Distinct Cellular Mechanisms

Liqun Yu1, Qing-yuan Huang1, Nelson Rebola4, Hai-Ying Shen1, Eric Kirsten Rapp1, Yuan-Ji Day3, Jarrod Ferrara1, Joana E. Coelho1, Paula M. Canas4, Zhi-Hong Huang2, Darcie Taylor1, Michael Moskowitz2, Michael Schwarzschild2, Joel Linden3, Rodrigo A. Cunha4 and Jiang-Fan Chen1

1Department of Neurology, Boston University School of Medicine, Boston, MA; USA 2Neuroscience Center and Department of Neurology, Massachusetts General Hospital, Boston, MA; USA 3Department of Internal Medicine, University of Virginia, Charlottesville, VA; USA 4Center for Neuroscience, Institute of Biochemistry, Faculty of Medicine, University of Coimbra, Portugal chenjf@bu.edu

The adenosine A2A receptor (A2AR) has recently emerged as a leading non-dopaminergic therapeutic target for Parkinson's disease for its ability to regulate motor activity. Furthermore, A2ARs influence brain injury outcome in variety of neurological disease models, presumably through modulation of glutamate release. Using forebrain neuronal-specific A2AR knockout (KO) mice, we here provide the first direct evidence that A2AR-mediated control of motor function and neuroprotection involve distinct cellular mechanisms. By crossing the floxed A2AR mice with the CaMKII-Cre transgenic line, we selectively depleted A2AR mRNA and protein in forebrain neurons to the background level of the global A2AR KO mice, as demonstrated by in situ hybridization, immunochemistry and receptor binding assays. This genetic deletion of A2ARs in forebrain neurons abolished the psychomotor effect of the A2AR selective agonist CGS21680 and antagonist KW-6002 and of the non-selective antagonist caffeine, and largely attenuated the pyschostimulant effect of cocaine. This demonstrates the key role of forebrain neuronal A2ARs in the modulation of psychomotor activity. In contrast, genetic deletion of the A2AR in forebrain neurons did not confer protection against ischemic brain injury by middle cerebral arterial occlusion or against MPTP-induced dopaminergic neurotoxicity, despite abolishing CGS21680-mediated presynaptic facilitation of glutamate release in forebrain A2AR KO mice. Furthermore, intracerebral ventricular administration of KW-6002 into forebrain A2AR KO mice reinstated neuroprotection against MPTP neurotoxicity. These results provide the clearest data yet that A2AR activity in forebrain neurons is critical to control psychomotor activity, but not for neuroprotection against brain injury, indicating that A2ARs modulate motor activity and brain damage by distinct cellular mechanisms. This opens up the new possibility of selectively manipulating A2AR's motor and neuroprotective effects by targeting different cellular elements.

Adenosine and Huntington's Disease

Szu-Yi Chou1,2, Ming-Chang Chiang1,3, and Yijuang Chern1,2,3,

1Institute of Biomedical Sciences, Academia Sinica, Nankang, Taipei 115, Taiwan 2Institute of Life Sciences, National Defense Medical Center, Taipei 104, Taiwan 3Institute of Neuroscience, National Yang-Ming University, Taipei 112, Taiwan bmychern@ibms.sinica.edu.tw

Huntington's disease (HD) is an autosomal dominant neurodegenerative disease characterized by chorea, dementia, and psychiatric symptoms. The causative mutation is a CAG trinucleotide expansion in exon 1 of the Huntingtin (Htt) gene. The normal Htt gene has 35 or fewer repeats in the N-terminal region, whereas the appearance of neurological symptoms is associated with 36 or more CAG repeats in the Htt gene. Among all brain areas, neurons within the striatum and cortex are most susceptible to HD. Besides the well-characterized neurological deficits, metabolic abnormalities have also been reported in HD. In addition to the brain, formation of Htt aggregates has also been documented in peripheral tissues including the liver. Several rodent models based on chemical lesions of the striatum or genetic manipulations of the Htt gene have been established and extensively used to develop beneficial treatments for HD. Among the four adenosine receptors, the A1 and A2A adenosine receptors have been suggested as potential drug targets in HD. Using an HD transgenic mouse model (R6/2), we found that daily administration of an A2A adenosine receptor-selective agonist (CGS21680, CGS) delayed the progressive deterioration of motor performance and prevented a reduction in brain weight. 3D-µMRI analysis revealed that CGS reversed the enlarged ventricle-to-brain ratio of R6/2 mice, with particular improvements in the left and right ventricles. 1H-MRS showed that CGS significantly reduced the increased choline levels in the striatum. Immunohistochemical analyses further demonstrated that CGS reduced the size of ubiquitin-positive neuronal intranuclear inclusions (NIIs) in the striatum of R6/2 mice and amended mutant Htt aggregation in a striatal progenitor cell line overexpressing mutant Htt with expanded polyQ. Moreover, chronic CGS treatment also ameliorated the formation of Htt aggregates and the decrease in two protein chaperones in the liver. Collectively, the above findings support the potential use of adenosine-related agents for treating HD.

Adenosine receptor-receptor interactions: Relevance for ischemia and caffeine tolerance

Rafael Franco, Sergi Ferré, Rodrigo Cunha, Vicent Casadó, Carmen Lluis, Francisco Ciruela

Molecular Neurobiology Unit. Institut d'Investigacions Biomédiques August Pi i Sunyer (IDIBAPS). University of Barcelona. Diagonal 645. 08028 Barcelona rfranco@ub.edu

Of the four known adenosine receptors (A1, A2A, A2B and A3), adenosine A1 receptors (A1Rs) and adenosine A2A receptors (A2ARs) are primarily responsible for the central effects of adenosine. In addition to their postsynaptic location in different brain regions, A1R and A2AR can be found presynaptically, where they modulate neurotransmitter release. A1R and A2AR are coupled to Gi/o and Gs/olf proteins, respectively. Presynaptic A1Rs are the prototype of G protein-coupled receptors whose stimulation decreases the probability of neurotransmitter release. On the other hand, presynaptic A2ARs are mostly excitatory and their stimulation induces neurotransmitter release. In this presentation evidence for heteromerization of A1 and A2A receptors is provided. This heteromerization have physiological consequences as described below.

Previous studies have provided evidence for colocalization and functional antagonistic interactions between A1Rs and A2ARs that modulate glutamate release in the striatum and hippocampus. A similar kind of interaction has been suggested to exist in the terminals of striatal cholinergic interneurons, although the localization and functional significance of A2AR in striatal cholinergic neurons is controversial. The coexistence of both stimulatory A2ARs and inhibitory A1Rs in the same terminal is intriguing particularly in view of their opposite functional effects.

A1R and A2AR form A1R–A2AR heteromers in transfected cells and in striatal glutamatergic nerve terminals. A1R–A2AR heteromerization plays a role in controlling the affinity of A1R and A2AR for agonists and antagonists, like caffeine. Functional studies show that A1R–A2AR heteromers are responsible for a strong A1RA2AR antagonistic cross_talk. The strength of this cross-talk would depend on the concentration of extracellular adenosine, which is enhanced in ischemia. Tolerance to caffeine can be explained by the occurrence of A1R–A2AR heteromers. Heteromerization of receptors for a given neurotransmitter or neuromodulator constitutes a novel way to regulate presynaptic neurotransmission.

Part of the results in this presentation will be published in 2006: Ciruela et al., J Neurosc. In the press.

Ectonucleotidases Physiological Implication

Biphasic regulation of airway E-NTPDases by Pseudomonas Aeruginosa lipopolysaccharide

Maryse Picher1 and Lauranell H. Burch2

1Cystic Fibrosis Center, University of North Carolina, North Carolina, USA 2Laboratory of Respiratory Biology, NIEHS, Durham, North Carolina, USA pichm@med.unc.edu

Chronic obstructive lung diseases are characterized by the inability to prevent bacterial infection causing recurrent inflammatory responses, epithelial damage and loss of lung function. In the past decade, numerous studies have demonstrated the importance of extracellular nucleotides for bacterial clearance. Their binding to P2 receptors on airway epithelia stimulates mucus secretion, cilia beating activity and hydration of the airway surface liquid layer. On the other hand, abnormally-high ATP levels resulting from damaged epithelia and bacterial lysis may cause lung edema and exacerbate inflammation-related epithelial damage. This study demonstrates the importance of ecto nucleoside triphosphate diphosphohydrolases (E-NTPDases; ATP → ADP → AMP) for the regulation of P2 receptor-mediated airway functions and identifies the mechanisms regulating their activities in lung diseases. All experiments were conducted on polarized primary cultures of human bronchial epithelial cells maintained at air-liquid interface. The E-NTPDase inhibitor, azide, reduced the metabolism of UTP and UDP by 45% and 55%, respectively. Chronic lung diseases, including cystic fibrosis (CF) and primary ciliary dyskinesia (PCD), exhibited 4–6 fold higher rates of nucleotide elimination and azide-sensitive E-NTPDase activities. The mechanisms regulating these ectonucleotidases (NTPDase 1 and NTPDase 3) under pathological conditions were first studied using the common lung pathogen, Pseudomonas aeruginosa lipopolysaccharide (LPS). Bronchial cultures exposed to an optimum (30 ng/ml) endotoxin concentration displayed a biphasic response characterized by an acute reduction in total activity (<8 h) followed by chronic (24 h) increases in activity, NTPDase 1 and 3 expression. P. aeruginosa LPS induces several acute responses from airway epithelia, including oxidative stress, lipid peroxidation and inflammation. Bronchial cultures were exposed (2 h) on the mucosal surface to LPS (30 ng/ml) or TNFα (10 ng/ml) in the absence/presence of bilateral scavengers of peroxide (catalase) or superoxide radicals (superoxide dismutase; SOD). Although total azide-sensitive NTPDase activity was reduced by acute exposures to LPS or TNFα, ebselen-sensitive NTPDase 3 activity was increased 3-fold while NTPDase 1 activity was completely inhibited. Normal activity levels were partially restored by the scavengers (catalase > SOD). The contribution of inflammatory pathways was demonstrated by the protective effect of the anti-inflammatory cytokine, IFNγ, and by SN50, cell permeable peptide preventing the translocation of the pro-inflammatory transcription factor, NFkB, to the nucleus. Altogether, these results suggest that NTPDase 1 and NTPDase 3 are co-expressed on human airway epithelia. Their opposite regulation by acute insults supports different roles in airway homeostasis. In contrast, elevated E-NTPDase activities in chronic lung diseases may represent an attempt to prevent P2 receptor desensitization and nucleotide-mediated lung damage.

This study was supported by the Cystic Fibrosis Foundation (Picher 05G0)

CD38/CD157 gene family in health and disease

Fabio Malavasi

Laboratory of Immunogenetics, Department of Genetics, Biology and Biochemistry, University of Torino Medical School, Torino, Italy fabio.malavasi@unito.it

The plasma membrane of human cells hosts a relatively large number (∼5%) of molecules acting as enzymes apparently operating in an anti-economical manner. Nucleotide-metabolizing ectoenzymes constitute a family within this larger family and are represented by a set of molecules involved in the catabolism and scavenging of extracellular nucleotides. This process results in the synthesis of compounds that play a critical role in cell homeostasis and metabolism, suggesting that the physiological role of this complex family goes beyond the simple recycling of nucleotides.

In this context, the story behind the CD38 gene family (CD38 and CD157) is representative of a more general trend involving several nucleotide-metabolizing ectoenzymes, such as CD39 and CD26. Indeed, human CD38 is the mammalian prototype of a family of phylogenetically conserved proteins which share structural similarities and enzymatic activities involved in the production of an intracellular second messenger with calcium mobilizing effects. Engagement of CD38 by agonistic monoclonal antibodies and the CD31 ligand initiates a cytoplasmic signaling cascade involving tyrosine phosphorylation of the proto-oncogene c-cbl and of the extracellular regulated kinase 1/2 complex. Further requirements for signal transduction include a privileged localization within the cholesterol-rich areas of the plasma membrane and physical association with specialized surface receptors. CD38-mediated signals are crucial in heterotypic cell adhesion and migration as well as in the activation of proliferation / survival programs by normal and neoplastic cells. A direct role in the pathogenesis of diseases is also suggested by the involvement of CD38 in directing the prognosis of chronic lymphocytic leukemia. Further studies concerning the association between the CD38 gene family and human diseases will provide a better insight into the biological functions of the complex family of receptors and enzymes.

Ectoenzymes of CD38 and CD39 gene families catalyze immunomodulation

Silvia Deaglio1,2, Karen M. Dwyer 2, David Friedman 2, Terry B. Strom 2,3 and Simon C. Robson2

1Department of Genetics, Biology and Biochemistry, University of Torino, Torino, Italy 2Departments of Medicine and 3 Surgery, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusettes, USA silvia.deaglio@unito.it

CD39 [ENTPD1 (ecto nucleoside triphosphate diphosphohydrolase-1), EC] is a member of a family of ecto-nucleotidases, surface molecules with extracellular catalytic sites that hydrolyze nucleotides. CD39 is the rate limiting enzyme that hydrolyses ATP/UTP and ADP/UDP to AMP; AMP is then rapidly degraded to adenosine by soluble or membrane-bound 5′-nucleotidase such as the ubiquitous CD73 (EC This family of nucleotide-metabolizing ectoenzymes also includes molecules for which a clear-cut role in lymphocyte activation has been extensively proven, such as CD26 and CD38. Specifically, experience gathered with the CD38 gene family has clearly shown a non-redundant role in leukocyte trafficking and migration to inflammatory sites. Further, human disease models such as chronic lymphocytic leukemia, have highlighted a crucial receptor-like role for CD38 in controlling proliferation and survival signals to normal and neoplastic lymphocytes.

Little is known on the role of CD39 in lymphocyte activation. Preliminary data indicates that the enzymatic functions controlled by CD39 are involved in the recruitment, activation and polarization of naive T cells by Langherhan_s dendritic cells. We have now expanded these observations by exploring CD39 expression and the functional role within quiescent and activated T cell subsets. Our data indicate that CD39 efficiently distinguishes Treg from other resting or activated T cells in humans and mice. Furthermore, CD39 serves as an integral component of the suppressive machinery of Treg, acting, at least in part, through the modulation of extracellular levels of adenosine. We show that the coordinated regulation of CD39 expression and of the adenosine receptor A2A activates an immuno-inhibitory network that regulates Th1 responses. The in vivo relevance of this network is manifest in the phenotype of Cd39 null mice. These mice spontaneously develop autoimmune diseases associated with heightened Th1 responses. These data confirm the potential of CD39 as a phylogenetically conserved marker of Treg that has immunomodulatory functions.

Finally, these results may also be read in the wider context of a network of ectoenzymes acting in synergy to scavenge nucleotides and to generate powerful mediators of immune responses. Further work will be needed to determine the functional interactions between this complex family of cell surface enzymes.

Role of CD73-derived adenosine in acute and chronic inflammation

Schrader J.1, Zernecke A.2, Buchheiser A.1, Weber Ch.2, Özüyaman B.1

1Department of Cardiovascular Physiology, University of Duesseldorf, Germany 2Institute of Molecular Cardiovascular Research, RWTH Aachen, Germany

We have recently reported that mice with targeted deletion of ecto-5′-nucleotidase/CD73 are characterized by reduced coronary flow, enhanced platelet activation and increased adherence of monocytes to the endothelium. In the present study we have investigated the molecular mechanisms by which this proinflammatory response is mediated and explored whether CD73-derived adenosine modulates neointima formation in an acute injury model and inhibits development of arteriosclerosis in ApoE/CD73−/− double mutants.

CD73−/− mice exibit increased luminal staining for vascular adhesion molecule (VCAM)-1 in carotid arteries and increased expression of VCAM-1 transcripts and protein in whole carotid lysates. Endothelial cells cultured from CD73−/− show an up-regulation of mRNA and protein expression of VCAM-1, which was associated with in creased nuclear factor (NF)-kB activity. Measurement of expression of the A1, A2a, A2b and A3 receptor by RTPCR in the aorta and freshly harvested aortic endothelial cells revealed substantial downregulation the A1 receptor (20% of control) while the other adenosine receptors remained unchanged. Ex-vivo perfusion of carotid arteries show that the incresed monocyte arrest in carotid arteries of CD73- is mediated by α4β1 integrin. After wire-induced injury of the carotid artery, CD73−/− expression was upregulated in WT mice, while neointima formation and macrophage content was increased in CD73−/− mice, concomitant with elevated NF-kB activation, luminal VCAM-1 expression and soluble VCAM-1 concentration. Treatment of mice with the specific A2a recetor agonist ATL-146e reversed the increased VCAM-1 transcript and protein expression in CD73−/−derived endothelial cells. Most importantly, ATL-145e fully prevented wire-induced neointima formation in CD73−/− mice.

To explore whether CD73-derived adenosine also modulates chronic inflammation we have generated ApoE/ CD73−/− double mutants and found in 6 month old animals kept on normal diet, that the development of arteriosclerotic lesions in the aorta was 2.5 fold higher compared with ApoE controls. Measurement of various cytokines in plasma substantiate the increased inflammatory state in the double knockout.

Our data demonstrate that CD73-derived adenosine through activation of A2a recetors protects against vascular inflammation, monocyte recruitment and neointima formation. This adenosine is also important in limiting the progression of arteriosclerosis.


Congenital and drug-induced P2Y12 defects in the clinical setting

Marco Cattaneo

Unità di Ematologia e Trombosi, Ospedale San Paolo, Dipartimento di Medicina, Chirurgia e Odontoiatria, Università di Milano, Milano, Italy marco.cattaneo@unimi.it

P2Y12 plays an important role in platelet function. It mediates ADP-induced platelet aggregation and amplifies platelet secretion induced by release-inducing platelet agonists. Congenital defects of the platelet P2Y12 receptor is characterized by mild/moderate bleeding tendency. Six patients with congenital P2Y12 defect, belonging to five kindred, have been described so far. Five patients had severe deficiency of the protein, while one expressed normal number of dysfunctional P2Y12 receptors on his platelets. P2Y12 is important in the pathogenesis of arterial thrombosis, since substances inhibiting its function are potent antithrombotic drugs. The thienopyridines ticlopidone and clopidogrel, which irreversibly inactivate the receptor, dramatically decrease the incidence of arterial thrombosis in patients at risk. The extent of inhibition by clopidogrel (and ticlopidine) of ADP-mediated platelet responses varies widely among individuals, and insufficient inhibition has been called “clopidogrel resistance”. Both ticlopidine and clopidogrel are a pro-drugs, which need to be metabolized by the liver to their active metabolites with anti-aggregating activity. Therefore, their pharmacological effects can be detected only some time after their first administration and, more importantly, the plasma levels of the active metabolites, and, consequently, the degree of inhibition of platelet aggregation induced by ADP, vary widely among subjects. In published studies, about 50% of the patients were either clopidogrel non responders or low responders. Interindividual variability in platelet inhibition by clopidogrel correlated well with the metabolic activity of the hepatic cytochrome P450, which activates the pro-drug to its active metabolite. Interference with clopidogrel metabolism by other drugs that are frequently given to patients with atherosclerosis, such as atorvastatin, can increase the number of patients who are resistant to clopidogrel, although this is still a controversial issue. A more recent thienopyridine, prasugrel, is currently under study. Preliminary data suggest that it may accomplish a greater inhibition of platelet aggregation than ticlopidine and clopidogrel and that the interindividual variability in response may be less marked. Cangrelor and AZD6140 are direct antagoinists of the platelet P2Y12 receptor. Both are currently under further clinical investigation.

P2Y1 and P2X1 as targets for new antiplatelet agents

C. Gachet

INSERM U311, EFS-Alsace, University Louis Pasteur, Strasbourg, France christian.gachet@efs-alsace.fr

ADP and ATP play a key role in normal haemostasis as well as in arterial thrombosis where platelet activation is pivotal. Upon activation, the nucleotides are released from the so-called dense granules, and stimulate three separate P2 receptors, the ATP-gated P2X1 cation channel and the two ADP receptors P2Y1 and P2Y12. The P2X1 receptor is involved in platelet shape change, activation by collagen and thrombus formation under high shear. The P2Y1 receptor initiates platelet aggregation through mobilisation of intracellular calcium stores while the P2Y12 receptor amplifies the responses to all agonists and stabilizes the aggregates. Among the platelet P2 receptors, the P2Y12 receptor is the molecular target of the efficient antiplatelet drug clopidogrel and several competitive antagonists which are under clinical evaluation. Its expression seems to be limited to the megakaryocytic lineage although several other cell types have been shown to express P2Y12 transcripts. It is thus a well established target for antiplatelet drugs. Less is known concerning the P2X1 and the P2Y1 receptors which are broadly expressed in many tissues. However, studies in knock-out mice and experimental thrombosis models using selective antagonists have shown that both the P2X1 and the P2Y1 receptors are potential targets for new antithrombotic drugs.

P2Y1 deficient mice display increased resistance to thromboembolism induced either by a mixture of collagen and adrenaline or by tissue factor. Moreover, IV administration to mice of the P2Y1 antagonists MRS2179 or MRS2500 similarly results in prolongation of the bleeding time, inhibition of ex vivo platelet aggregation to ADP and resistance to thromboembolism. Models of localized arterial thrombosis have also been used to further evaluate the P2Y1 receptor as a target for antiplatelet therapy. In a model of superfusion of ferric chloride to induce injury of mesenteric arterioles and intravital microscopy examination it could be shown that P2Y1 inhibition was as efficient as P2Y12 inhibition. Interestingly, the treatment of P2Y1 deficient mice by clopidogrel resulted in additive inhibition of thrombosis, suggesting that a combination of drugs targeting both P2Y1 and P2Y12 could be of benefit. A second model of localized arterial thrombosis consists in laser injury of mesenteric arterioles. In this case, P2Y1 deficient mice or MRS2500 treated mice again displayed reduced thrombosis as compared to the wild type but to a lesser extend as compared to mice treated with maximal doses of clopidogrel. Again, the combination of MRS2500 with clopidogrel resulted in additive inhibition of thrombosis. P2X1 deficient mice similarly display resistance to all these experimental thrombosis models, including systemic thromboembo- lism and localized thrombosis. Moreover, the suramin analog NF449 which displays some selectivity for P2X1 as been shown to reduce thrombosis in vivo.

Altogether, these results indicate that both P2Y1 and P2X1 are potential targets for new antiplatelet compounds, which should be evaluated now in experimental thrombosis in higher animals.

Purinoceptor-evoked signalling via P2X1 and other ion channels in the platelet

Martyn Mahaut-Smith1, Gwen Tolhurst1, C.Y. Eleanor Fung1, Richard N. Carter1, Catherine Vial2, Catherine Leon3, Christian Gachet3, and Richard J. Evans2

1Department of Physiology, University of Cambridge, Cambridge, CB2 3EG, UK 2Department of Cell Physiology & Pharmacology, University of Leicester, Leicester, LE1 9HN, UK 3Institut National de la Sante et de la Recherche Medicale (INSERM) U.311, EFS-Alsace 10, Strasbourg Cedex, 67065, France mpm11@cam.ac.uk

The role of ion channels in platelet function has proven difficult to investigate due to the challenge of conducting direct electrophysiological recordings from these small and fragile cell fragments. As the megakaryocyte must generate most, if not all proteins of the anuclear platelet, the giant precursor cell provides a useful means to study the regulation and identity of platelet ion channels. Indeed, using confocal fluorescence imaging of fibrinogen binding to murine megakaryocytes, we show that ADP-evoked activation of αIIbβ3 requires co-stimulation of P2Y1 and P2Y12 receptors as is well established in platelets. Under whole-cell patch clamp, hexokinase-purified ADP stimulated multiple phases of inward currents, which conduct both Na+ and Ca2+ into the cell. All currents were dependent upon the presence of P2Y1 receptors, although P2Y12 receptors played a synergistic role through a PI 3-kinase-dependent pathway. The main conductance showed both early transient and delayed sustained phases, which mirrored the intracellular Ca2+ increase. In addition, P2X1 receptor currents appeared as multiple transient currents, indicating that ATP secretion stimulates this non-selective cation channel in a repetitive, focal manner. Transient receptor potential (TRP) ion channels represent candidates for the direct P2Y receptor-dependent current and screening by RT-PCR from individually selected megakaryocytes indicates that TRPC6 is the dominant message amongst the TRPs known to be stimulated by phospholipase-C activity. To examine the relative role of P2X1 receptors as an autocrine pathway for Ca2+ influx in human platelets, we established pharmacological conditions that selectively block this pathway (NF449 or α,βmeATP predesensitisation). Multiple platelet agonists utilised the P2X1 receptor as a major pathway by which they increase [Ca2+]i. The maximal reduction of peak [Ca2+]i increases following P2X1 inhibition was 48.3 ± 1.3%, 75.5 ± 8.8%, 65.3 ± 3.1% and 30.9 ± 6.6% for thrombin (0.03 Uml−1), collagen (0.5 µg ml−1), U46619 (thromboxane A2 analogue, 1.0 µM) and ADP (30 µM), respectively. We conclude that P2X1 receptors represent an important and widespread means of elevating Ca2+ in the platelet but that P2Y1/12 receptors stimulate another non-selective cation channel, most likely TRPC6, with a role in ADP-evoked Ca2+ influx.

Supported by the British Heart Foundation, Wellcome Trust and Medical Research Council

Nucleotide Receptors and Cellular Functions

Cardiovascular ATP release: Focus on the red blood cell

David Erlinge

Department of Cardiology, Lund University, Sweden

Previous studies have demonstrated that nucleotides can be released from various sources in the cardiovascular system such as platelets, sympathetic nerves, endothelial cells and cardiomyocytes. We have focused on the regulation of nucleotide release in man, and the possible clinical implications.

ATP is released from red blood cells (RBC) when intracellular cAMP levels are increased in response to reductions in oxygen tension and pH. The released ATP then binds to P2Y receptors on the endothelium and stimulates vasodilatation. Because blood consists of approximately 40% RBCs, containing a 1000-fold higher ATP concentration than plasma (mM vs. µM), even a minor release of ATP from the high intracellular concentrations could have major circulatory effects and may need to be regulated. We found that the ATP degradation product ADP inhibits ATP release by acting on P2Y13 receptors that inhibit cAMP in the RBC [1]. This negative feedback system could be important in the control of plasma ATP levels and tissue circulation. The first selective P2Y13 antagonist MRS2211, blocked the ADP induced inhibition of ATP release in micromolar concentrations, giving us a tool to evaluate the importance of P2Y13 negative feedback pathway for circulatory control.

Nucleotide release in the heart comes from both cardiac myocytes and endothelial cells, but the RBC may also be an important source. In patients with acute myocardial infarction, we found that both ATP and UTP are released and may have effect on cardiac function and coronary flow [2]. The ATP receptor P2Y11, the UTP receptor P2Y2 and the UDP receptor P2Y6 are expressed in human myocardium and mediate inotropic effects [2, 3]. The prominent increase in blood flow following ischemia (reactive hyperemia) is to a major part mediated by ADP acting on endothelial P2Y1 receptors [4].

In the peripheral vasculature we found that purines and pyrimidines may provide a link between glucose and diabetic microvascular disease [5]. High glucose induced NFAT activation in cerebral microvessels by a mechanism totally blocked by apyrase and by 50% by the P2Y6 receptor blocker MRS2578, indicating that the vascular growth factors ATP, UTP and UDP are released in response to high glucose and may stimulate intracellular reactions associated with microvascular disease.Inflammation in the atherosclerotic plaque is important for destabilisation and plaque rupture that is the major cause of myocardial infarction. We have indirect evidence that ATP is released in the plaque. The common Ala-87-Thr polymorphism of the ATP receptor P2Y11 that is present in one fifth of the population was examined in 1200 patients with myocardial infarction and 2400 controls and markedly increased the risk of myocardial infarction, especially for early myocardial infarction and patients with family history. Interestingly, the inflammatory marker C-reactive protein was significantly increased in carriers of the Ala-87-Thr polymorphism, indicating an inflammatory mechanism.

In conclusion, purines and pyrimidines seem to be released and involved in most aspects of human cardiovascular pathophysiology.

Joachim Jankowski; Charite, Berlin “Uridine adenosine tetraphosphate: a novel endothelium- derived vasoconstrictive factor”

J. Jankowski

Charitè, Berlin

Beyond serving as a mechanical barrier, the endothelium shows important regulatory functions. The discovery of nitric oxide (NO) revolutionized our understanding of vasoregulation. In contrast, the identity of endothelialderived vasoconstrictive factors (EDCFs) remains unclear. The supernatant obtained from mechanically stimulated human endothelial cells obtained from dermal vessels elicited a vasoconstrictive response in an isolated perfused rat kidney. Endothelial-derived vasoconstriction in the isolated perfused rat kidney was decreased by a purinoceptor blocker more than by an endothelin receptor blocker. The nucleotide, uridine adenosine tetraphosphate (Up4A) was isolated from the supernatant of stimulated human endothelium and identified by mass spectrometry. Up4A most likely exerts vasoconstriction predominantly via P2X1 receptors, and probably also via P2Y2 and P2Y4 receptors. In healthy subjects Up4A plasma concentrations are found which cause vasoconstriction. Stimulation with adenosine 5′-triphosphate (ATP), uridine 5′-triphosphate (UTP), acetylcholine, endothelin, A23187 and mechanical stress release Up4A from endothelium, suggesting that Up4A contributes to vascular autoregulation. Up4A is the first dinucleotide containing both purine and pyrimidine moieties isolated from living organisms. We conclude that Up4A is a novel potent non-peptidic EDCF. Its vasoactive effects, plasma concentrations and its release upon endothelial stimulation strongly suggest that Up4A has a functional vasoregulatory role.

P2Y and P2X receptors and their role in exocrine pancreas

Ivana Novak

August Krogh Building, Institute of Molecular Biology and Physiology, University of Copenhagen, Copenhagen, Denmark inovak@aki.ku.ak

The exocrine pancreas secretes digestive enzymes and fluid rich in HCO3and poor in Ca2+. These secretions originate from acini and ducts in response to classical secretagogues cholecystokinin and secretin. The function of these epithelia needs to be coordinated, such that digestive enzymes are not prematurely activated or that calcium stones do not form. We propose that along the acini-duct axis ATP and other components of the purinergic cascade mediate short-term regulation of secretion. The following findings lay the basis for these proposals. Acini release ATP into the interstitium and most importantly into the lumen leading to pancreatic ducts [1]. The initial ATP concentrations are in the high miromolar range, but since secretion also contains specific nucleotidases, CD39 and CD73, these would regulate concentration of nucleotides/nucleosides along the ductal tree [2].

Acini themselves are relatively poor in functional P2 receptors. In contrast, pancreatic ducts, which secrete HCO3rich fluid, express several types of functional P2 receptors on the luminal and basolateral membranes [3]. On cellular level, P2X4 and P2X7 receptors stimulate Na+ and Ca2+ influx, but seem to be without prominent effects on Cl and H+/ HCO3 transport usually associated with fluid secretion [3, 4]. However, possibly via ERK1/ERK2 signalling pathways, they could be modulators of secretion. Initial studies on mice stimulated with pilocarpine in vivo indicate that exocrine gland function is affected in P2X7 knockout animals.

Pancreatic ducts also express metabotrophic P2Y2 and P2Y4 receptors, which in many epithelia, including human pancreatic duct cell lines, can stimulate Ca2+ activated Cl channels [5]. In native rat pancreatic ducts, patch-clamp experiments indicated inhibition of K+ conductance [3]. In a recent study we found that pancreatic ducts expresses Ca2+ activated K+ channels of a big conductance (BK, maxi-K+ channels) and intermediate conductance (IK). We used the oocytes expression and electophysiological recordings to study the interaction between individual P2Y receptors and Ca2+-activated K+ channels. The most interesting result was that stimulation of P2Y2 receptors inhibited BK [6]. Thus, as in the native epithelium, P2Y2 receptors inhibit BK, which would down-regulate secretion, possibly preventing over-distension of the duct. In summary, elucidation of P2X and P2Y receptor regulated processes in pancreatic ducts will provide a better understanding of epithelial secretion in health and disease, such as in cystic fibrosis.

The projects were supported by the Danish Medical and Science Research Councils and the Lundbeck Foundation.

Purinergic signalling triggers development of the eye

Nicholas Dale

Department of Biological Sciences, University of Warwick, Coventry, UK N.E.Dale@warwick.ac.uk Abstract not received

Round Table on Purinome-based drugs

Allosteric activators of glucokinase: The development of cyclopropane-derived agents and their potential for use in the treatment of type two diabetes

Jochen Ammenn1, David G. Barrett1, Martina Berg1, Martin B. Brenner1, Stephen L. Briggs2, Annie Delaunois3, Jim D. Durbin2, Alexander M. Efanov1, Ulrich Giese1, Gema Sanz Gil1, Jesper Gromada1, Haihong Guo2, Ulrike Hary1, Stefan Heuser1, Astrid Kahl1, Carsten Ott1, Mark Radloff1, Rainer Riedl1, Ulrike Roettig1, Horst Schwetzler1, Erich Seger1, Sabine Sewing1, Birgit Sommer1, Yong Wang1, Jutta Wanner1, Andreas Weichert1, Andrea Zaliani1, and Christian Zechel1

1Lilly Research Laboratories, 22419 Hamburg, Germany 2Lilly Research Laboratories, Indianapolis, Indiana 46285 3Lilly Development Centre, 1348 Mont-Saint-Guibert, Belgium barrett_david_g@lilly.com

The glucose-sensing enzyme glucokinase (GK) plays a key role in glucose metabolism. A member of the hexokinase family, it catalyzes the first step in glycolysis, phosphorylation of glucose to glucose 6-phosphate. It is a unique hexokinase in that it has a low affinity for glucose, demonstrates non-Michaelis-Menten kinetics and displays no inhibition by the product of the reaction [1]. These particular features make GK an ideal sensor of physiological changes in blood glucose levels, translating them into changes in the metabolic status of the cell.We report here the effects of a novel small molecule glucokinase activator, LY2121260, which enhances GK activity via binding to an allosteric site located in the hinge region of the enzyme [2]. We discuss the affect of LY2121260 on the glucose affinity of the enzyme and the velocity of the reaction, and present crystallographic data that support its mechanism of action. We show that LY2121260 stimulates insulin secretion in pancreatic â -cells and increases glucose use in rat hepatocytes in a glucose-dependent manner. Furthermore, animals treated with LY2121260 show an improved glucose tolerance following an oral glucose challenge. We also describe some structure-activity studies within this class of GK activators. Our results support the concept that GK activators increase both insulin secretion and hepatic glucose use and in doing so may prove to be effective agents for the control of blood glucose levels in patients with type 2 diabetes. Open image in new window

Beyond purinergic receptors: Drugging the rest of the purine-binding proteome

S. Hall, A. Barabasz, T. Barta, J. Daw, L. Dubois, J. Eaves, P. Fadden, B. Foley, T. Freed, L. Geng, G. Hanson, K. Hardemann, L. Hinkley, M. Jenks, M. Hu, K. Huang, M. Lewis, L. Liu, W. Ma, J. Otto, J. Partridge, B. Pronk, J. Rice, A. Scott, M. Silinski, P. Steed, J. Veal, K. Verleysen, R. Ware, and T. Wadkins

Serenex, Inc. 323 Foster Street, Durham, NC, 27701 USA shall@serenex.com

Purinergic receptors represent an attractive class of druggable targets within the purine-binding proteome, yet there are many other enzyme families of interest as therapeutic targets within this large superfamily. We estimate that the purine-binding proteome contains at least 2000 members, with kinases composing the largest fraction(25%) of this group. We report here a novel approach to screen hundreds of these targets in parallel, thus providing an efficient way to probe compound libraries as ligands to a broad variety of enzyme classes. We constructed and screened a targeted library of 8000 compounds using a novel affinity displacement assay (referred to as proteome mining). Multiple tissues were used as the source of native purine-binding proteins to profile this library. Hits were obtained for nearly 100 targets, including well validated targets such as dihydrofolate reductase and multiple kinases (e.g. CDK2, Her2, PDK1, AurA), and less well-validated targets such as the kinases Fer and Nek9. Importantly, the method also provided affinity information on potential toxicity targets (e.g. phosphorylase kinase, pyruvate carboxylase, glucose-6-phosphate dehydrogenase). Thus, this approach provides primary target affinity information as well as selectivity data across the entire purine-binding proteome. In addition, affinities of compounds for their protein targets could be determined directly from the same assay and these measurements were sufficiently quantitative and reproducible to drive the development of structure-activity relationships. Unexpectedly, selectivity within a gene family was not predictive of selectivity between gene families across the purine-binding proteome. The application of this screening approach to four diverse targets, dihydrofolate reductase, Hsp90, quinone reductase 2, and PI3kinase will be described. In each case, the use of the affinity displacement assay enabled the identification of novel, selective, orally active compounds. Taken together, these results support the use of proteome mining as both an alternative to single target biochemical screens and as a tool to drive SAR development. Open image in new window

Challenges and Opportunities in the Identification of Insulin-like Growth Factor-I Receptor Kinase Inhibitors

C. García-Echeverría, M.A. Pearson, T. Meyer, J. Mestan, J. Zimmermann, J. Brueggen, H.-G. Capraro, R. Cozens, D.B. Evans, D. Fabbro, P. Furet, D. Graus Porta, J. Liebetanz, G. Martiny-Baron, S. Ruetz, S. Jacob, and F. Hofmann

Novartis Institutes for BioMedical Research, CH-4002 Basel, Switzerland carlos.garcia-echeverria@novartis.com

The Insulin-like Growth Factor-I Receptor (IGF-IR) is a member of the insulin receptor family of tyrosine kinases. This transmembrane-spanning protein is composed of two α- and two β-subunits linked by disulfide bonds. While the !-subunits are extracellular, the β-subunits span the plasma membrane and encompass an intracellular tyrosine kinase domain devoted to the initiation of several signal transduction cascades. Signaling through IGF-IR is initiated upon binding of the cognate ligand—Insulin-like growth factor-I (IGF-I) or II (IGF-II)—to the extracellular domain of the receptor. It is thought that this peptide-protein interaction induces a conformational change that results in auto-transphosphorylation of each β-subunit at specific tyrosine residues within the intracellular kinase domain and outside the catalytic domain. Activation of the receptor triggers, through docking and/or phosphorylation of several transduction molecules (e.g., IRS-1, IRS-2 or Shc), results in activation of downstream signaling pathways, of which the Ras/Raf/MAPK pathway is primarily responsible for mitogenesis, and the survival PI3K/PKB pathway appears to play a major role in mediating the IGF-IR biological functions.

A broad range of experimental studies have revealed that IGF-IR function is implicated in most of the hallmarks of cancer, but it is probably the anti-apoptotic activity of this receptor that makes its kinase activity an attractive therapeutic target in anti-cancer drug discovery. In this context, the identification of specific lowmolecular mass inhibitors of IGF-IR has proven to be a major challenge for medicinal chemistry due to the high sequence identity at the kinase domains of IGF-IR and InsR (around 84%) and, in particular, at the ATP-binding pocket (100% sequence identity). Selectivity over the InsR is a critical requirement for any IGF-IR kinase inhibitor to avoid disturbing glucose homeostasis. This presentation will cover the identification and characterization of a new series of IGF-IR kinase inhibitors. When used alone or in combination with cytotoxic agents, the synthetic IGF-IR kinase inhibitors suppress tumor growth and prolongs survival of mice with diffuse bone lesions of multiple myeloma. These preclinical findings and additional studies support the potential application of targeted therapeutic strategies directed at IGF-IR in combination with established antitumor modalities or for the treatment of IGFs-responsive neoplasias. Moreover, the selectivity achieved at the cellular level with these IGF-IR inhibitors suggest conformation differences between the native forms of IGF-IR and InsR -from the unactivated to the fully activated form—that can effectively be exploited for drug discovery.

Kinases: Important drug targets for novel anti-inflammatory therapies

Alan J. Lewis

Novacell, 31 Technology Drive, Suite 100, Irvine, CA 92618sides”” alan.lewis@hotmail.com

Protein kinases contribute to a diverse number of cellular processes and are the second most important group of drug targets after G-protein-coupled receptors. Moreover, aberrant kinase activity is implicated in many human diseases in particular those cancer and inflammatory diseases. Of the more than 500 kinases in the human kinome very few are targeted by currently marked drugs, notably the anticancer agents Gleevec (BCR Abl kinase, c-Kit), Tarceva (EGFR) Sorafenib (Raf, VEGFR) and Sutent (EGFR) Oriology remains the most fertile area for kinase inhibitors followed by inflammatory diseases. One reason is that many kinase inhibitors are multi-targeted (“dirty drugs”) and the lack of selectivity may be beneficial in cancer to prevent resistance from occurring.

Several kinases including p38MAPK, JNK and IkB kinase-2 (IKK2) have been identified as key targets for a plethora of inflammatory diseases and clinical trials are underway. A major advantage in directing antiinflammatory drug discovery programs against such targets is the ability to modulate multiple inflammatory gene products including cytokines such as TNF and IL-1, cell adhesion molecules as well as enzymes such as Cox-2 and MMPs that are pivotal in the pathology of inflammatory disease.

Many of the current kinase inhibitors were developed from “focused” small molecule kinase libraries that target the ATP-binding site and the availability of crystal structures for several key kinases has greatly facilitated lead optimization. Allosteric strategies have also been used successfully to identify kinase inhibitors notably p38. The family of human protein kinases accounts for most cellular signal transduction and can provides an enormous opportunity to develop new treatments for anti-inflammatory therapies.


[3H]Adenine's High Filter Binding Precludes its Use as a Radioligand for Adenine Receptors

Kai Ye, Thea Mulder-Krieger, Margot W. Beukers and Ad P. Ijzerman

Division of Medicinal Chemistry, Leiden/Amsterdam Center for Drug Research, Leiden University, PO Box 9502, 2300 RA Leiden, The Netherlands k.ye@lacdr.leidenuniv.nl

An orphan G protein-coupled receptor, rat MrgA10, one of the MAS-related G protein-coupled receptors, has recently been characterized as an adenine receptor [1]. Among the methods used in the de-orphanization process were binding studies with [3H]adenine as a radioligand [1]. In subsequent studies from other research groups saturation analysis, kinetic studies and displacement experiments were performed on native tissues such as rat brain cortex with [3H]adenine as a radioligand, demonstrating the presence of adenine binding sites in these tissues [2, 3].

We initiated a program to address the medicinal chemistry aspects of the adenine receptor, with radioligand binding studies as a primary screen. We prepared membranes from CHO cells stably transfected with the rat adenine receptor (rAR) kindly donated by IriDM (Scherpenheuvel, Belgium), next to membranes of rat brain cortex. Subsequently, radioligand binding assays with [3H]adenine as the radioligand were developed. Open image in new window

In that process we observed a number of worrying phenomena. Firstly, we observed an unusually rapid association of the radioligand in our kinetic studies, consistent with literature [2]. Secondly, in the displacement assays we learned that decreasing the membrane protein concentration (see Figure for rat brain cortex membranes) led to an increase in “specific” [3H]adenine binding, such that in the absence of protein the highest binding was observed. Thirdly, the affinities of adenine and a number of adenine derivatives to compete for [3H]adenine filter binding were in good agreement with the values in literature for the adenine receptor [1, 2, 3].

In conclusion, radioligand binding studies aimed at the adenine receptor with [3H]adenine as the radioligand should be performed with great care, if at all.

[3H]HEMADO—A New Tritiated Adenosine Receptor Agonist Radioligand

Karl-Norbert Klotz1, Nico Falgner1, Sonja Kachler1, Rosaria Volpini2, Diego Dal Ben2, Catia Lambertucci2, Ram Chandra Mishra2, Sauro Vittori2 and Gloria Cristalli2

1Universität Würzburg, Institut für Pharmakologie und Toxikologie, Versbacher Str. 9, D-97078 Würzburg, Germany and 2Università di Camerino, Dipartimento di Scienze Chimiche, Via S. Agostino, 1, I-62032 Camerino, Italy klotz@toxi.uni-wuerzbvurg.de

A3 adenosine receptors are promising drug targets for a number of conditions like inflammatory diseases including asthma, ischemic injury or certain types of cancer. Consequently, intense efforts in many laboratories are dedicated to the development of A3 selective agonists adn antagonists. A number of radioligands are in use for screening of new compounds in binding assays. In agonist development it is advantageous to use an agonist as the radioligand in order to avoid biphasic competition curves as a high affinity radiolabeled agonist will bind to high affinity binding sites only. In the case of A3 adenosine receptors the two radioligands 125I-AB-MECA and [3H]-NECA are widely used. A tritiated ligand offers a number of advantages in routine applications. However, [3H]NECA is of limited value owing to the relatively low affinity for human A3 receptors (Ki-value 6.2 nM), lack of subtype selectivity and its tendency to bind to numerous proteins other than adenosine receptors.

Based on a recently characterized series of potent 2-substituted adenosine receptor agonists we developed a tritiated A3 selective radioligand. The ligand of choice was HEMADO (2-hexyn-1-yl-N6-methyladenosine) which exhibits an Ki-value at the human A3 subtype of 1 nM. HEMADO is 300fold selective versus the A1 subtype and 1,100fold selective compared to the A2A receptor [1]. Starting from the precursor 2-hexyn-1-yl-6-iodopurine-9-riboside [3H]HEMADO was generated by substitution of the 6-iodo with tritiated methylamine. The resulting radioligand has an specific radioactivity of 890 MBq/mmol (24 Ci/mmol). Open image in new window

Characterization of [3H]HEMADO in radioligand binding studies revealed reversible binding to the human A3 adenosine receptor. In saturation binding studies for the A3 subtype a KD value of 1.3 nM was determined. The nonspecific binding at a radioligand concentration of 0.5 nM amounted to 1–2% of total binding. Competition binding with a panel of agonists and antagonists clearly confirmed the correct A3 pharmacology of the binding site labeled by [3H]HEMADO. In concentrations up to 10 nM no specific binding was detectable at any of the other adenosine receptor subtypes.

With [3H]HEMADO we present a tritiated high affinity agonist with ≥300fold A3 selectivity and very low nonspecific binding. [3H]HEMADO is a useful tool for specific screening for A3 receptor agonists and antagonists in radioligand binding assays.

[3H]-MRE 2029-F20, A Novel Selective Antagonist Radiolig and, for the Pharmacological and Biochemical Characterization of A2B Receptors

Stefania Gessi1, Katia Varani1, Stefania Merighi1, Elena Cattabriga1, Youri Szabadkai2, Rosario Rizzuto2, Karl Norbert Klotz3, Edward Leung4, Stephen Mac Lennan4, Pier Giovanni Baraldi5 and Pier Andrea Borea1

1Dpt of Clin. and Exp. Med., Pharmacology Unit and “Interdisciplinary Center for the Study of Inflammation (ICSI),” 2Gen. Pat. Unit and 3Universitat Würzburg, Germany; 4King Pharmaceuticals Research & Development, Cary, North Carolina and 5Dpt of Pharm. Sci., University of Ferrara, 44100 Italy gss@unife.it

This work investigates the pharmacological and biochemical properties of A2B adenosine receptors in hA2BHEK293 cells and in peripheral blood cells by using a new potent 8-pyrazole xanthine derivative, [3H]-N-benzo[1,3]dioxol-5-yl-2-[5-(1,3-dipropyl-2,6-dioxo-2,3,6,7-tetrahydro-1H-purin-8-yl)-1-methyl-1H-pyrazol-3-yl-oxy]-acetamide] ([3H]-MRE 2029-F20), that shows high affinity and selectivity for hA2B versus hA1, hA2A and hA3 subtypes. In particular MRE 2029-F20 displays low affinity for the human A1 receptor (Ki = 245 ± 31 nM) and no significant affinity for the human A2A and A3 subtypes (Kis >1,000 nM). [3H]-MRE 2029-F20 bound specifically to the hA2B receptor stably transfected in HEK293 cells with KD of 2.8 ± 0.2 nM and Bmax of 450 ± 42 fmol/mg protein. Saturation experiments of [3H]-MRE 2029-F20 binding in human neutrophils and lymphocytes detected a single high affinity binding site with KD of 2.4 ± 0.5, 2.7 ± 0.7 nM and Bmax of 79 ± 10, 54 ± 8 fmol/mg of protein, respectively, in agreement with real-time RT-PCR studies showing the presence of A2B mRNA. The rank order of potency of typical adenosine ligands with recombinant hA2B receptors was consistent with that typically found for interactions with the A2B subtype and was also similar in peripheral blood cells. NECA stimulated cAMP accumulation in both hA2BHEK293 and native cells whereas phospholipase C activation was observed in recombinant receptors and endogenous subtypes expressed in neutrophils but not in lymphocytes. MRE 2029-F20 revealed to be a potent antagonist in counteracting the agonist effect in both signal transduction pathways. In conclusion [3H]-MRE 2029-F20 is a selective and high affinity radioligand for the hA2B adenosine subtype and in this work it has been used to compare the presence and functional coupling of A2B receptor in a recombinant system and in peripheral blood cells.

2,6,8-Trisubstituted-1-Deazapurines as Adenosine A1 Receptor Antagonists

Jacobien K. von Frijtag Drabbe Künzel, Lisa C. W. Chang, Thea Mulder-Krieger, Joost Westerhout, Thomas Spangenberg, Johannes Brussee and Adriaan P. IJzerman

Leiden/Amsterdam Center for Drug Research, Division of Medicinal Chemistry, P.O. Box 9502, 2300 RA Leiden, The Netherlands ijzerman@lacdr.leidenuniv.nl

Exploration of two particular pyrimidine and purine series in previous publications [1, 2] led to a refinement of a pharmacophore defined for antagonists of the adenosine A1 receptor. This contribution details the adoption of these new criteria to produce a series of 1-deazapurines with consistently high affinity for the adenosine A1 receptor. 1-Deazapurines (otherwise known as 3H-imidazo[4,5-b]pyridines) are structurally very similar to the purines, however, in a synthetic sense they posed an array of challenges, mainly as a result of the reduced reactivity about the 6-membered ring. The desired double aromatic substituents at the 2- and 6-positions were amongst the most troublesome features to incorporate. An eventual adaptation of a known route resulted in a series with five of the derivatives displaying Ki values in the sub-nanomolar range. Thus LUF5983 displayed an affinity of 0.87 nM at the human adenosine A1 receptor with >200-fold selectivity vs. human A2A and A3 receptors. The compound was shown to behave as a potent antagonist/inverse agonist in cAMP second messenger studies in cells expressing the human adenosine A1 receptor. Open image in new window

2-Arylpyrazolo[3,4-c]quinolin-4-Heteroaroylamines as Potent and Selective Human A3 Adenosine Receptor Antagonists. Synthesis and Biological Evaluation

Daniela Catarzia, Vittoria Colottaa, Flavia Varanoa, Francesca Capellia, Ombretta Lenzia and Katia Varanib

aDipartimento di Scienze Farmaceutiche, Università di Firenze, via U. Schiff 6, 50019, Sesto Fiorentino (FI). bDipartimento di Medicina Clinica e Sperimentale, Sezione di Farmacologia, Universitàdi Ferrara, via Fossato di Mortara 17–19, 44100 Ferrara daniela.catarzi@unifi.it

Adenosine mediates a wide variety of physiological effects by interacting with four receptor subtypes: A1, A2A, A2B and A3. All four adenosine receptor (AR) subtypes are coupled via G-protein to the adenylate cyclase in either an inhibitory (A1 and A3) or stimolatory manner (A2A and A2B) [1]. AR antagonists have attracted great attention for their potential therapeutic use, and the A3 AR antagonists are sought as potential anti-inflammatory and anti-asthmatic agents [2]. In the last few years in our laboratory, much effort has been directed toward the synthesis of tricyclic heteroaromatic systems rationally designed as antagonists of the hA3 AR subtype. In a recent paper we reported the study on 2-arylpyrazolo[3,4-c]quinolin-4-amines which were endowed with nanomolar hA3 affinity [3]. On this class of derivatives, the presence of acyl residues on the 4-amino group produces profitable effects both in terms of hA3 affinity and selectivity [3, 4]. On this basis, we decided to synthesize a new set of 2-arylpyrazolo[3,4-c]quinolin-4-amines in which diverse heteroaroyl moieties were introduced on the 4-amino group. Open image in new window

Affinities of the new derivatives at hA1, hA2A and hA3 AR and their inhibitory effects on NECA-stimulated cAMP production in hA2B CHO cells were determined. The obtained results show that the 2-arylpyrazolo[3,4-c]quinoline-4-heteroaroylamines are potent and selective hA3 antagonists, indeed they possess hA3 nanomolar affinities while they are completely inactive at the other three AR subtypes.

2-Arylpyrazolo[3,4-c]quinolin-4-Acylamines as Potent and Selective Human A3 Adenosine Receptor Antagonists. Synthesis, Biological Evaluation and Molecular Modeling Studies

Vittoria Colotta1, Daniela Catarzi1, Flavia Varano1, Francesca Capelli1, Ombretta Lenzi1, Letizia Trincavelli2 and Stefano Moro3

1Dipartimento di Scienze Farmaceutiche, Università di Firenze, via U. Schiff 6, 50019, Sesto Fiorentino (FI).

2Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Università di Pisa, via Bonanno 6, 50126 Pisa. Molecular Modeling Section, Dipartimento di Scienze Farmaceutiche, Università di Padova, via Marzolo 5, 35131 Padova. vittoria.colotta@unifi.it

Adenosine is a ubiquitous neuromodulator eliciting many biological effects by activation of four different receptors subdivided into A1, A2A, A2B and A3 subtypes and belonging to the G protein-coupled receptor family [1]. In the last decade much research has been directed towards the study of adenosine receptor (AR) antagonists which have proved to be attractive therapeutics in many pathophysiological conditions [2]. In particular, the A3 AR subtype antagonists are putative anti-inflammatory, anti-asthmatic or anti-ischemic agents [2]. A part of our recent studies were aimed at finding new tricyclic A3 AR antagonists and among them the 2-arylpyrazolo[3,4-c]quinolin-4-amino series was investigated [3]. In the preliminary study a single modification at a time was apported at the R1 or R4 position of 2-phenylpyrazolo[3,4-c]quinolin-4-amino scaffold and some of the mono-substituted derivatives resulted to be potent human (h) A3 AR antagonists. SAR studies pointed out that methyl and methoxy groups at meta or para positions were the most profitable R1 substituents for enhancing hA3 AR affinity. Advantageous effects were also elicited by introducing an acetyl or a benzoyl group at the R4 position. These interesting results prompted us to continue investigating this class of hA3 AR antagonists with the purpose of evaluating whether the contemporary functionalization of the key positions R1 and R4 with the above cited groups could increase hA3 AR affinity and selectivity. Also the bulkier diphenylacetyl substituent or two benzoyl residues were probed on the 4-amino group. Open image in new window

Preliminary binding data at the hA3 AR show that the simultaneous presence of the suitable substituents on the 2-phenyl ring (R1) and the 4-amino group (R4) maintained the hA3 AR affinity in the nanomolar range. Molecular modeling studies are in progress in order to elucidate the binding mode of this series of derivatives to the hA3 receptor site and to rationalize the affinity data.

2-Chloro-2′-Deoxyadenosine-Induced Apoptosis of Human Cancer Cells Bearing a Mutated p53 Isoform: Role of MAP Kinases

Stefania Ceruti, Alessia Mazzola and Maria P. Abbracchio

Laboratory of Molecular and Cellular Pharmacology of Purinergic Transmission — Department of Pharmacological Sciences, University of Milan — via Balzaretti 9 — 20133 — Milan, Italy stefania.ceruti@unimi.it

The biochemical pathways controlling cell death are often mutated and/or defective in cancer cells, contributing to the development of resistance to chemotherapy. Thus, for any given anti-cancer agent, it is important to verify whether it can effectively induce cell death in cancerous cells carrying mutations of various apoptotic pathways. We have previously demonstrated that human astrocytoma ADF cells are resistant to agents activating the intrinsic pathway of apoptosis due the presence of a mutated caspase-9 isoform (1). Nevertheless, the anti-cancer agent 2-chloro-2′-deoxyadenosine (Cladribine, 2CdA) induces cell cycle block and is still able to effectively kill these cells by recruiting an atypical pathway of cell death, with caspase-2 acting as an “initiator” caspase, followed by caspase-3 (2). To better characterize 2CdA-activated apoptotic pathway in ADF cells, in this study we have focussed our attention on the tumor suppressor protein p53 which is mutated in more than 50% of human cancers. By cloning and sequencing the p53 isoform expressed by ADF cells, we detected a single nucleotide mutation in position 797 which translated into a G-to-E substitution at position 266 which belongs to the DNA-binding domain of the p53 protein. Since phosphorylation of the Ser15 residue of the p53 protein represents one of the most frequent post-translational modification controlling its proapoptotic activity, we have analyzed the importance of this phosphorylation reaction in cells exposed for various time periods to 2CdA by means of a specific antibody directed against phosphorylated p53. A very rapid apperance of the protein band corresponding to phospho-Ser15 p53 was observed, starting from a 30 min and declining after 3 hours of incubation with the adenosine analog. P53 phosphorylation was completely prevented by 2-deoxy-cytidine, which competes with 2CdA for its intracellular phosphorylation to the corresponding chloro-deoxy nucleotides. Indeed, co-incubation with SP600125, a selective inhibitor of c-jun-terminal kinases 1 and 2 (JNK1/2), completely blocked p53 phosphorylation, whereas inhibitors of other members of the MAP kinase family of enzymes (e.g., ERK1/2 and p38) had no effect. SP600125 was also able to significantly reduce the percentage of apoptosis induced by 2CdA. However, p53 phosphorylation and 2CdA-induced apoptosis are not correlated to each other, since α-pifithrin, which selectively inhibits p53 transcriptional activity downstream of post-translational modifications, was completely ineffective in preventing cell death. Taken together, these results suggests that a JNK-dependent but p53-independent pathway of death is recruited by 2-CdA. JNK1/2 were also involved in the 2CdA-induced cell cycle block at G0/G1 phases and enhanced expression of genes controlling cell cycle progression (e.g., p21, GADD45A) as demonstrated by microarray analysis. Despite the previously mentioned inability of the selective inhibitor PD098059 to prevent p53-phosphorylation, we have also detected a pro-apoptotic role for ERK1/2, since PD098059 significantly reduced 2CdA-mediated cell death. These results suggest that 2CdA treatment might represent an effective pharmacological approach for cancer cells bearing mutations of different apoptotic pathways of death, by recruting various members of the MAP kinases family.

2′-C-Methyl Derivatives of Tecadenoson and 2-Chloro-Tecadenoson with Increased Selectivity for Human A1 Adenosine Receptor

Cappellacci L,1 Franchetti P,1 Vita P,1 Petrelli R,1 Pasqualini M,1 Costa B,2 Spinetti F,2 Martini C 2, Klotz K-N,3 Lavecchia A,4 and Grifantini M1

1Department of Chemical Sciences, University of Camerino, 62032 Camerino, Italy; 2Department of Psychiatry, Neurobiology, Pharmacology and Biotechnology, University of Pisa, 56126 Pisa, Italy; 3Institut für Pharmakologie, Universitat Würzburg, Würzburg, Germany, 4Department of Medicinal Chemistry and Toxicology, University of Naple “Federico II,” 80131 Naple, Italy loredana.cappellacci@unicam.it

Adenosine mediates its physiological effects through four G protein-coupled receptors (A1, A2A, A2B and A3). A large number of agonists with high affinity at A1, A2A, A3 adenosine receptors and moderate affinity at A2B receptor have been developed over the years. Many compounds originally thought to be selective for the A1 or A2A subtypes later turned out to be also potent agonists at the more recently discovered A3 receptor. Owing to the great interest in A1 agonists as neuroprotective, antilipolytic, anti-arrhythmic, and antinociceptive agents, there is a need for novel agonists with high potency and selectivity at this receptor subtype to avoid side effects due to the stimulation of the other subtypes. We discovered that the substitution of the hydrogen in 2′-position of the ribose moiety of the A1 selective agonist CCPA with a methyl group (2′-Me-CCPA) reduces the affinity at human A2A and A3 receptors, thus increasing the selectivity for A1 subtype [1]. In this communication we report on the affinity at human adenosine receptors of 2′-C-methyl derivatives of Tecadenoson (CVT510, N6-3(R)-tetrahydrofuranyl-adenosine), a potent and selective A1 agonist in clinical trials for treatment of supraventricular tachyarrhythmias, and of its 2-chloro derivative (2-Cl-Tecadenoson). Tecadenoson, 2-Cl-Tecadenoson and their 2′-C-methyl derivatives were evaluated for the affinity at human A1, A2A, A2B, and A3 ARs expressed in CHO cells in comparison with CPA, CCPA, 2′-Me-CPA and 2′-Me-CCPA. With the exception of Tecadenoson that proved to be equipotent to CPA, N6-3(R)-tetrahydrofuranyl substituted derivatives showed only 2.5- to 5-fold less affinity at human A1 AR in comparison with the corresponding N6-cyclopentyl analogues. It was further confirmed that the introduction of a methyl group in the 2′-position of the ribose moiety of adenosine analogues induces a relevant decrease of affinity at A2A and A3 receptors. Thus, 2′-Me-Tecadenoson and 2′-Me-2-Cl-Tecadenoson turned out to be very selective compounds for human A1 AR (>4,762- to >6,452-fold vs human A2A, and 211- to 288-fold vs A3 receptor, respectively).

2′-C-Methyl-2-Chloro-N6-Cyclopentyladenosine, A Potent and Highly Selective A1 Adenosine Receptor Agonist, Has Antinociceptive Activity and Modulates RVM On- and Off-Cell Activities

de Novellis V.,1 Mariani L.,1 Vita D.,1 Giordano C.,1 Cappellacci L.,2 Franchetti P.,2 Grifantini M.,2 Rossi F.,1 and Maione S.1

1Department of Experimental Medicine, Second University of Napoli, 80138 Napoli, Italy 2Department of Chemical Sciences, University of Camerino, 62032 Camerino, Italy

This study was undertaken to investigate the effect of 2′-C-methyl-2-chloro-N6-cyclopentyladenosine (2′-Me-CCPA), a potent and highly selective A1 adenosine receptor agonist [1], on nociceptive responses and on spontaneous activity of ON- and OFF-neurons of rostral ventromedial medulla (RVM). Moreover, we investigated whether the intra-periaqueductal grey (PAG) microinjection of 2′-Me-CCPA induced changes in the tail flick latencies, as well as tail flick-related changes in RVM cell activities.

Systemic administrations of 2′-Me-CCPA (2–5 mg/kg, i.p.), 10 min before formalin, reduced the late hyperalgesic phase; in formalin test, these effects were prevented by DPCPX (3 mg/kg, i.p.), a A1 receptor antagonist, but not by DPMX (3 mg/kg, i.p.), a A2 receptor antagonist.

Intra-PAG microinjections of 2′-Me-CCPA (0.5–1 nmol/rat) increased the tail flick latencies, delayed the tail flick-related onset to ON-cell burst and decreased the duration of OFF-cell pause in a dose dependent manner. Furthermore, 2′-Me-CCPA decreased RVM ON-cell and increased OFF-cell ongoing activities, in a dose dependent manner. These electrophysiological effects were prevented by DPCPX (1 nmol/rat.).

In conclusion, this study confirms the role of A1 receptors in the modulation of inflammatory pain and suggests a critical role of PAG purinergic system for the control of acute and chronic pain.

3-Modified 2-Aminothiophenes as A1AR Allosteric Enhancers

Peter J. Scammells1, George Nikolakopoulos1 and Joel Linden2

1Department of Medicinal Chemistry, Victorian College of Pharmacy, Monash University, 381 Royal Parade, Parkville VIC 3052 Australia and 2Departments of Medicine (Cardiology) and Pharmacology, University of Virginia, Charlottesville, VA 22908, USA peter.scammells@vcp.monash.edu.au

The first allosteric enhancers (AEs) acting at the A1 adenosine receptor (A1AR) were reported by Bruns et al. in 1990 [1, 2]. PD81,723 (1) was one of the more potent and effective of the initial series of enhancers and has subsequently been commonly used for benchmarking new AEs. Since this initial discovery, other researchers have directed significant effort to refining the structure-activity relationships of the 3-, 4- and 5-positions of the 2- aminothiophene core. Open image in new window

However, relatively few 2-aminothiophenes with esters or amides in the 3-position have been tested as allosteric enhancers. The initial study by Bruns and co-workers included two 2-amino-3-ethoxycarbonyl-4,5,6,7-tetrahydrothieno[2,3-c]pyridines [1, 2]. Whilst these compounds maintained some enhancing ability, they proved to be less potent than the corresponding 3-benzoyl substituted 2-amino-4,5,6,7-tetrahydrothieno[2,3-c]pyridines. We recently synthesised a series of substituted 2-amino-3-benzoyl-4,5-diphenylthiophene allosteric enhancers. En route to these target compounds we also prepared some novel 2-aminothiophene-3-carboxylates and carboxamides to explore the effect of these substituents on AE activity [3]. Three of the five alkoxycarbonyl compounds proved to be clearly more effective than PD81,723 (1), with 3-trifluoromethylbenzyl 2-amino-4-(4-methylphenyl)-5-phenylthiophene-3-carboxylate (2) being the most effective compound in this group. We now report a more complete structure-activity study of this class of compound as A1AR allosteric enhancers. A number of compounds, notably 3, proved to be more potent and efficacious than PD81,723 (1). Conformationally restricted analogs have also been prepared and evaluated.

A C-terminal Lysine that Controls Human P2X4 Receptor Desensitization

Samuel J. Fountain & R. Alan North

Faculty of Life Sciences, Michael Smith Building, University of Manchester, Manchester, M13 9PT, UK. Samuel.fountain@manchester.ac.uk

Receptor desensitization can determine the time course of transmitter action, and profoundly alter sensitivity to drugs. Among P2X receptors, ion currents through homomeric P2X4 receptors exhibit intermediate desensitization when compared to P2X1 and P2X3 (much faster) and P2X2 and P2X7 (slower) [1]. We recorded membrane currents in HEK293 cells transfected to express the human P2X4 receptor. The decline in current during a 4 s application of ATP (100 µM) was about 30%; this was not different during whole-cell or perforated patch recording. Alanine-scanning mutagenesis of the intracellular C-terminal identified two positions with much accelerated desensitization kinetics (Lys373: 92 ± 3.2% and Tyr374: 74 ± 4.2%). At position 373, substitution of Arg or Cys also strongly accelerated desensitization: however, in the case of K373C the wild-type phenotype was fully restored by adding ethylammonium methanethiosulfonate. At position 374, phenylalanine could replace tyrosine. These results indicate that wild type desensitization properties requires an aromatic moiety at position 374, and an amino rather than a guanidino group at position 373. These residues lie between previously identified motifs involved in membrane trafficking (YXXXK and YXXGL) [2, 3], and implicates the C terminal also in rearrangements leading to channel closing during the presence of agonist.

5-Heteroarylcarbamoylamino-Pyrazolo-TriazoloPyrimidines as hA3 Adenosine Receptor Antagonists

G. Pastorin1, T. Da Ros1, C. Bolcato1, C. Montopoli2, S. Moro2, B. Cacciari3, P. G. Baraldi3, K. Varani4, P. A. Borea4 and G. Spalluto1

1Dipartimento di Scienze Farmaceutiche, Università degli Studi di Trieste; 2Molecular Modeling Section, Dipartimento di Scienze Farmaceutiche, Università degli Studi di Padova; 3Dipartimento di Scienze Farmaceutiche, Università degli Studi di Ferrara 4Dipartimento di Medicina Clinica e Sperimentale-Sezione di Farmacologia, Università degli Studi di Ferrara; Italy Spalluto@units.it

A new series of pyrazolotriazolopyrimidines bearing at the N5-position different heteroarylcarbamoylamino moieties are described as an enlarged series of previously reported highly potent, selective and water soluble human A3 adenosine receptor antagonists (1) [1]. Open image in new window

The synthesized compounds showed A3 adenosine receptor affinity in the nanomolar range and different levels of selectivity evaluated in radioligand binding assays at human A1, A2A, and A3 adenosine receptors. In particular, the effect of the heteroaryl substituents at the N5 position has been analyzed and this study allows recognizing that the presence of a pyridinium moiety in this position not only increases water solubility but also improves or retains potency and selectivity at the human A3 adenosine receptors. In contrast, replacement of pyridine with different heterocycles produces loss of affinity and selectivity at the human A3 adenosine receptors. A molecular modeling study has been carried out with the aim to explain these various binding profiles [2].

A Cell-based Assay Suitable for HTS of Adenosine A1 Receptor Agonists

Giuliana Piazza, Lia Scarabottolo

Axxam S.r.l., San Raffaele Biomedical Science Park, Via Olgettina, 58, 20132 Milan, Italy giuliana.piazza.gp@axxam.it

Purinergic receptors are a family of seven ubiquitous transmembrane receptors comprising two classes, P1 and P2, which are activated by adenosine and extracellular nucleotides, respectively. The P1 receptors signal through multiple intracellular effectors in response to nucleoside activation. There are four subtypes of P1 receptors: A1, A2A, A2B and A3.

In this study, we have focused our attention on the development of a cell-based system suitable for the identification of agonists of the adenosine A1 receptor, suitable for High Throughput Screening applications (HTS).

The A1 receptor is coupled to Gαi, which inhibits adenylyl cyclase. Additionally, it mediates the activation of several types of K+-channels, the inactivation of N-, P- and Q-type Ca2+ channels and the activation of phospholipase Cβ (via βγ subunits).

Mainly through the A1 receptor, adenosine mediates potent effects on myocardium and conduction tissue, endothelium and vascular smooth muscle, inducing delayed tolerance to ischemia-reperfusion injury.

In particular, A1-receptor activation produces preconditioning to protect the heart and other tissues from subsequent ischemic injury. Therefore, the pharmaceutical industry is interested in producing selective A1-receptor agonists as pharmacological treatment for cardiovascular diseases.

We have created a CHO cell line with stable expression of the reporter luciferase under the control of cAMP response elements and a chimera of the human A1 receptor and Gαs, which allows the switching of the receptor-mediated signal towards adenylil cyclase stimulation. We have demonstrated for the first time that activation of the A1 receptor can be detected by the direct increase of luminescence.

This generated A1-Gαs-expressing cell line is a sensitive and reliable functional cell-based assay suitable for the HTS of specific receptor agonists.

A Negative Inotropic Effect of Up4A in the Human Heart

Gergs U,1 Pönicke K,1 Simm A,2 Silber RE,2 Jankowski J, 2 Jankowski V, 2 Tepel M,2 Zidek W, Neumann J,

Institute for Pharmacology and Toxicology, Halle, 1Department of Cardio-Thoracic Surgery, Halle, 2Medizinische Klinik IV, Charité Campus Benjamin Franklin, Berlin Joachim.neumann@medizin.uni-halle.de

Adenosine has been reported to exert negative inotropic effects in isolated cardiac preparations of numerous mammalian species, including man. The effects are assumed to be mediated via an A1 adenosine receptor. Interestingly, derivatives of adenosine like diadenosine polyphosphates e.g. Ap4A exert similar effects via the A1 adenosine receptor in the human heart. Recently, uridine adenosine tetraphosphate (Up4A) was described as a novel endothelian derived vasoconstrictive factor (Nat Med 11:223, 2005). Up4A was released under certain conditions from endothelial cells and exerted potent vasocontrictory effects (EC50≅1 nM) via P2-receptors. Here, we studied the effect of Up4A on force of contraction in isolated electrically driven atrial preparations from human hearts. The preparations were obtained from patients undergoing cardiac bypass surgery due to coronary heart disease. Preparations were studied under isometrical conditions driven at 0.5 Hz.

Concentration response curves were obtained for the β-adrenoceptor agonist isoproterenol (1 nM–10 µM) and Up4A (1 nM–0.3 µM). In the very same preparations, isoproterenol exerted a positive inotropic effect while Up4A exerted a concentration dependent negative inotropic effect. The effect started at about 0.03 µM and amounted to 35% at 0.3 µM, the highest concentration of Up4A tested (n = 3). We assume that Up4A exerts the effect probably via A1 adenosine receptors. In summary, the new endogenous vasoconstrictor Up4A can reduce force of contraction in the human heart.

Hence, it may play an autoregulatory function in the cardiovascular system.

A Putative Dictyostelium Discoideum P2X Receptor

Melanie Ludlow & Steve Ennion Cell Physiology and Pharmacology, University of Leicester, England, LE1 9HN ml119@le.ac.uk

Completion of the Dictyostelium discoideum genome sequencing project has allowed a family of five putative P2X receptor genes to be provisionally identified in this species of cellular slime mould. Functional characterisation of a P2X or P2X-like receptor in this simple eukaryotic model organism could potentially contribute to our understanding of this class of ligand-gated ion channel. Extracellular ATP release from Dictyostelium and ATP induced calcium influx has previously been reported [1] supporting the presence of purinergic signalling pathways in this species.

The five putative Dictyostelium P2X subunits possess 35–75.8% sequence identity between each other and 11.4–16.1% sequence identity to human P2X1–7 subunits with conservation of key P2X receptor elements, such as an Nterminal PKC motif, a C-terminal membrane stabilisation motif, two potential hydrophobic transmembrane domains and a similar overall size (368–407 amino acids in length). The Dictyostelium gene showing most homology to vertebrate P2X channels (dP2XE) was selected for further investigation.

Initially we attempted to express dP2XE protein in Xenopus laevis oocytes as this expression system has been used successfully for the characterisation of vertebrate and Schistosoma mansoni [2] P2X receptors. Expression of dP2XE protein in oocytes was achieved following introduction of a vertebrate Kozak sequence to the N-terminus. However, cell surface biotinylation revealed that the protein was not trafficked to the plasma membrane in these cells and no ion channel currents could be detected by two-electrode voltage-clamp experiments. In the presence of a food source Dictyostelium exist in a unicellular state. Upon starvation, cells aggregate and develop into a spore containing fruiting body. Northern blot analysis of dP2XE expression demonstrated that the dP2XE gene is transcribed at a constant level through all stages of development, potentially indicating a housekeeping role. In order to determine the subcellular localisation of dP2XE in Dictyostelium, a C-terminal eGFP fusion protein (dP2XE-eGFP) was expressed. This protein displayed a diffuse cytoplasmic distribution with distinct localisation to the membranes of an intracellular organelle but absence from the plasma membrane. Confocal imaging of cells stained with the styryl dye FM2-10 [3] demonstrated localisation of dP2XE-eGFP to the membranes of contractile vacuoles, intracellular organelles involved in water expulsion/osmoregulation and potentially calcium regulation. This intracellular localisation of dP2XE indicates a novel function compared to that of the conventional cell surface membrane localised P2X receptors identified in vertebrates.

A Role for the P2X7 Receptor in Control of Toxoplasma gondii Infection

S.J. Fuller1, M. Lees2, H. Murray1, R. Sluyter1, B. Gu1, N. Boulter2, K. Skarratt1, N.C. Smith2, J.S. Wiley1

1The University of Sydney, Department of Medicine, Nepean Hospital, Penrith 2750 and 2Institute for the Biotechnology of Inectious Diseases, University of Technology, Sydney sfuller@med.usyd.edu.au

The P2X7 receptor has a role in the host immune response to infection with the obligate intracellular pathogens, Mycoplasma and Chlamydia. We describe three immunocompetent subjects with absent P2X7 function due to one or a combination of inherited loss-of-function single nucleotide polymorphisms all of whom had a prolonged, severe illness due to infection with the obligate intracellular protozoan parasite, Toxoplasma gondii. T gondii usually causes an asymptomatic infection in immunocompetent humans but infection in immunocompromised individuals can cause serious disease, including encephalitis while intrauterine infection can cause congenital blindness, mental retardation and death. Our first subject was a 20-year-old female who presented with submandibular lymphadenopathy. Serology was positive for recent Toxoplasma infection and a lymph node biopsy showed features consistent with toxoplasmosis. The second subject was a 24-year-old female who had a foetal ultrasound scan at 22 weeks gestation that showed gross cerebral ventriculomegaly consistent with congenital Toxoplasmosis and confirmed by PCR on amniotic fluid. The third subject, a 14-year-old male, presented with a 2-year history of tiredness, lethargy and generalized painless lymphadenopathy. Serology for T gondii was positive for acute infection and an excision biopsy of an enlarged left axillary lymph node showed lymphadenitis consistent with toxoplasmosis. Function of the P2X7 receptor, as measured by ethidium uptake using time resolved flow cytometry, was absent in two subjects and low in the third. Sequencing of the P2X7 gene confirmed the presence of one or a combination of loss-of-function single nucleotide polymorphisms previously described by our group. All subjects were HIV negative and had normal T-cell subsets and normal serum immunoglobulins Therefore, normal P2X7 function may be of importance in controlling Toxoplasma infection. Possible P2X7 dependent mechanisms of the host immune response to Toxoplasma infection include P2X7 dependent phagosome-lysosome fusion via phospholipase D activation and P2X7 triggered apoptosis of infected host cells, both of which have been shown to play a role in the host response to Chlamydia and Mycobacteria infection.

A Role for Nucleotide Signaling Pathways in Adult Neurogenesis

S.K. Mishra1, V. Shukla1, N. Braun1, C. Schomerus2, H.-W. Korf2, H. Kettenmann3, C. Gachet4, J. Sévigny5, and S.C. Robson6, Y. Ikahara7 and H. Zimmermann1

1Frankfurt University, Biocenter, 2Frankfurt University, Medical School, Germany, 3Max Delbrück Center for Molecular Medicine Berlin-Buch, Germany, 4INSERM U.311, EFS-Alsace, Strasbourg, France, 5Sainte-Foy, Québec, Canada, 6Harvard Medical School, Boston, USA, 7Fukuoka University, School of Medicine, Japan h.zimmermann@cns.uni-frankfurt.de

In the adult rodent brain, active neurogenesis takes place in the subventricular zone of the lateral ventricles (SVZ) and in the dentate gyrus of the hippocampus. In both neurogenic regions, astrocyte-like cells function as stem cells/precursor cells. To date the cellular and molecular events driving a subpopulation of precursor cells into neurogenesis as well as the cellular transition states leading to mature neurons are poorly understood. We showed that stem cells of the adult SVZ and progenitor cells in the dentate gyrus (residual radial glia) highly express the extracellular nucleoside triphosphate-hydrolyzing enzyme NTPDase2 [1]. Patch-lamp analysis of cells in hippocampal slices derived from mice transgenic for GFP under the control of the nestin promotor suggest that progenitor cells express P2X nucleotide receptors [2]. Progenitor cell-containing neurospheres cultured from the adult mouse SVZ in the presence of epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) express the ecto-nucleotidases NTPDase2 and the tissue non-specific isoform of alkaline phosphatase, hydrolyzing extracellular ATP to adenosine. ATP, ADP and to a smaller extent UTP evoke rapid Ca2+ transients in neurospheres that are exclusively mediated by the metabotropic P2Y1 and P2Y2 nucleotide receptors. In addition, agonists of P2Y1 and P2Y2 receptors and low concentrations of adenosine augment cell proliferation in the presence of the mitogenic growth factors EGF and bFGF. Neurosphere cell proliferation in the presence of growth factors is attenuated after application of the P2Y1 receptor antagonist MRS2179 and in neurospheres from P2Y1 receptor knockout mice. This suggests that constitutive P2 receptor activation via endogenously released nucleotide is necessary for optimal cell proliferation [3]. Our results infer nucleotide receptor-mediated synergism that augments growth factor-mediated proliferation of adult neural progenitor cells. They support the notion that extracellular nucleotides contribute to the control of adult neurogenesis in both, the SVZ and the dentate gyrus.

A SNP in the Adenosine A1 Receptor Associated with Fibromyalgia

Hailing Liu1, Bruce Solitar1, Jessica Stein1, Roland Staud2, Daniel Clauw3, Bruce Cronstein1

1New York University School of Medicine, New York, NY. 2University of Florida School of Medicine, Gainesville, FL. 3University of Michigan, Ann Arbor, MI.


Fibromyalgia (FM) is a chronic pain syndrome of unknown etiology characterized by widespread chronic pain and tender points. Although the exact pathophysiology of FM is unknown, individuals with FM are hyperalgesic to painful stimuli. Mice with targeted deletion of the adenosine A1 receptor (A1AR), like FM patients, demonstrate increased sensitivity to painful stimuli, and are characteristically anxious and aggressive. We therefore hypothesized that FM may be due to genetically-determined changes in A1AR.

Methods and Results

DNA was isolated and purified from peripheral blood from FM patients and age- and sexmatched normal controls. Initially the gene (promoters, exons, UTR) was sequenced in 14 FM patients and 24 controls. A single nucleotide polymorphism in the 3′UTR (C2526T) was observed in 3/14 FM patients and no controls. For all subsequent genotyping assays a 241 bp region containing the SNP was amplified and sequenced for all patients and controls. Patient samples and controls were obtained from two FM clinics and an increased frequency in the C2526T SNP was observed in the FM patients as compared to controls (Table below). To determine whether the C2526T SNP was in linkage equilibrium with other SNPs in the A1 gene the frequency of 3 known SNPs upstream and 3 SNPs downstream of the C2526T SNP was determined in patients and controls by Taqman Assay and no differences were observed (data not shown). The overall frequency of the C2526T polymorphism in the FM patients was 17.3% as compared to 6.8% in the controls (p = 0.019)



Frequency of C/T

































* p=0.019

Previous reports have suggested that subgroups of patients with FM have SNPs in the serotonin receptor gene (5HT2AR) and COMT, an enzyme involved in metabolism of adrenergic agents but we observed no difference between controls and FM patients with regard to the frequency of these SNPs (Data not shown)


A SNP in the A1AR (C2526T) is associated with FM suggesting that in some patients genetically determined changes in adenosine receptor expression or function lead to fibromyalgia.

A Truncated P2X7 Receptor Variant (P2X7-j) Endogenously Expressed in Cervical Cancer Cells Antagonizes the Full-Length P2X7 Receptor through Hetero-Oligomerization

Ying-Hong Feng MD PhD1, Xin Li MD PhD2, Liqin Wang MD1, Lingying Zhou MD2, George Gorodeski MD PhD2,3

Department of Pharmacology1, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Reproductive Biology2, Physiology and Biophysics3, and Oncology{ur3}, CASE (Case Western Reserve) University, Cleveland, Ohio, USA gig@cwru.edu

In human cervical epithelial cells the P2X7 receptor mediates ATP-induced apoptosis by a mechanism that involves pore formation, augmented calcium influx, and calcium-dependent activation of the apoptotic mitochondrial pathway. However, the degree of ATP- and BzATP-induced apoptosis is smaller in cancer than in normal cervical cells [1]. A PCR product of a smaller size than the expected full-length P2X7 was identified in experiments trying to amplify the full-length P2X7 gene. DNA sequencing and gene analysis of the smaller PCR product revealed an identical P2X7 gene that lacks exon 8 except the A882, with a shift of coding frame to a new variant. The predicted truncated naturally occurring variant (P2X7-j) (Genebank Accession Number DQ399293) is a polypeptide of 258 amino acids that lacks the entire intracellular carboxy terminus, the second transmembrane domain, and the distal third of the extracellular loop of the full-length P2X7 receptor. The P2X7-j cDNA was subcloned into various expression vectors and introduced into the host cells MDCK and HEK293 by transfection, and translation of a specific product (a protein cluster of 45–42 KDa) was confirmed. Expression of the P2X7-j as a naturally occurring protein was also confirmed in the human cancer cervical cell lines CaSki, HT3, SiHa, and Hela by Western blot with anti-P2X7 antibody. The P2X7-j was expressed mainly in the plasma membrane but compared to the full length receptor some of the P2X7-j also localized in nuclear/perinuclear regions. In host cells, inducible expression of full length P2X7 alone, but not P2X7-j alone resulted in significant decrease in cell number and increased apoptosis. In contrast, P2X7-j expressing cells showed an increase in cell number and diminished baseline and BzATP-induced apoptosis. Similarly P2X7-j expression showed diminished ligand-binding and channel function capacities and failed to form pores in response to treatment with BzATP. Co-expression of the P2X7-j did not significantly affect expression of the full length P2X7-j, but it inhibited / blocked BzATP-induced P2X7 channel activation, pore formation and apoptosis. The present results support the hypothesis that full length P2X7 molecules homo-oligomerize as oligo-trimers. Interestingly, the P2X7-j interacted with the P2X7-j in a manner suggesting P2X7-j/P2X7-j homo-oligomerization, and with the full-length P2X7 suggesting P2X7-j/P2X7 heterooligomerization. P2X7 immunoreactivities in HEK293 cells co-expressing the P2X7 plus the P2X7-j suggested formation of inactive complexes of the P2X7 receptor in the order of [P2X7-j]⋙[P2X7-j]2/[P2X7] > [P2X7]2/[P2X7-j]. These results identify a novel P2X7 variant with apoptosis-inhibitory actions, and demonstrate a distinct regulatory property for a truncated variant to antagonize its full-length counterpart through hetero-oligomerization. This may represent a general paradigm for regulation of a protein function by its variant. In cancer cells, hetero-oligomerization of the P2X7-j with the P2X7 could result in inactive receptor, and abrogated P2X7-mediated apoptosis could play a role in cervix neogenesis and enhance the growth of the cancer cells. Support: AHA-SDG-0030019N, NHLBI-HL65492 (YHF); NIH HD29924, AG15955 (GIG).

A1 Adenosine Receptor Antagonist, L-97-1, Increases Survival in Rats with Gram-Negative Septicemia

Masakatsu Goto1, Constance O. Vance2, and Constance N. Wilson2

1Department of Surgery, Loyola University of Chicago, Stritch School of Medicine, Maywood IL, 2Endacea, Inc., Research Triangle Park, NC. cwilson@endacea.nctda.org

Septicemia is a medical syndrome characterized by an overwhelming systemic response to infection that can rapidly lead to shock, organ failure and death. Following its release from Gram-negative bacteria, elevated levels of endotoxin in the plasma (endotoxemia) causes acute organ damage, including acute lung injury (ALI) and death. Previously, it was reported that A1 adenosine receptor (AR) antagonists block endotoxin-induced ALI and that endotoxin binds to and activates A1ARs on human pulmonary artery endothelial cells to induce the release of IL-6 and TXA2 [1, 2]. L-97-1 is a water-soluble, small molecule A1 AR antagonist with high affinity and high selectivity for the human A1 AR [3]. In a rat cecal ligation perforation (CLP) model of Gram-negative septicemia and endotoxemia, the effect of L-97-1 on mortality at 7 days (with and without antibiotics) following CLP was investigated. Administration of L-97-1 as an intravenous (i.v.) therapy post-CLP improved 7 day mortality in a dose-dependent manner (30–40% survival) as compared to untreated CLP control (17% survival) or antibiotics alone (23% survival). In combination with antibiotics (gentamicin plus ampicillin), L-97-1 increased survival to 50–70% in a dose-dependent manner. In this animal model of Gram-negative septicemia and endotoxemia, L-97-1 as an adjuvant therapy to antibiotics is more effective than antibiotics alone to improve 7 day mortality. Despite the introduction of an FDA approved adjuvant therapy, Xigris™ (drotrecogin alfa), the mortality rate remains high for sepsis. As an adjuvant therapy to antibiotics, L-97-1 may improve the outcome for patients with Gramnegative septicemia.

Supported by STTR Phase I Grant # AI056603

A1 Adenosine Receptor Antagonist versus Montelukast Sodium on Airway Reactivity and Inflammation

Ahmed Nadeem1, Peter C.M. Obiefuna1, Constance N. Wilson2 and S. Jamal Mustafa1

1Department of Pharmacology, East Carolina University, Greenville, NC;2 Endacea Inc, Research Triangle Park, NC, USA cwilson@endacea.nctda.org

Adenosine, an important signaling molecule in asthma, produces bronchoconstriction in allergic rabbits, primates, and humans by activating A1 adenosine receptors (ARs) [1, 2, 3]. The effects of activation of A1 ARs on airway inflammation in asthma are unknown. The effects of L-97-1, a water-soluble, small molecule A1 AR antagonist, are compared to montelukast sodium (MK), a cysteinyl leukotriene-1 (CysLT1) receptor antagonist on early and late phase allergic responses (EAR and LAR), bronchial hyper-responsiveness (BHR) to histamine, and airway inflammation following House Dust Mite (HDM) administration in a hyper-responsive rabbit model of allergic asthma. Rabbits were made allergic by intraperitoneal injections of HDM extract within 24 h of birth followed by booster HDM injections. Hyper-responsive rabbits were subjected to aerosolized HDM (2500 AU), 1 h after intragastric administration of L-97-1 (1 mg/kg) or MK (0.15 mg/kg) and, lung dynamic compliance (Cdyn) and airway inflammation was assessed. Cdyn was significantly higher following treatment with L-97-1 at all time points and with MK at 60–300 min following HDM (p<0.05); thus, L-97-1 blocks both EAR and LAR, whereas MK blocks only LAR. Both L-97-1 and MK significantly blocked BHR at 24 h (p<0.05). Both L-97-1 and MK significantly reduced bronchoalverolar lavage (BAL) eosinophils at 6 h and neutrophils at 6 and 24 h (p<0.05). As opposed to MK, L-97-1 significantly reduced BAL lymphocytes at 6 and 24 h, while MK significantly reduced BAL macrophages at 6 and 24 h (p<0.05). By blocking both bronchoconstriction and airway inflammation, L-97-1 may be an effective oral anti-asthma treatment.

Supported by North Carolina Biotechnology Center Kenan Award CFA #2000 CFG 8002 and STTR Phase I Grant # HL070458.

A2A Adenosine Receptor Down Regulation Leads to Increased Pro-Inflammatory Gene Expression in A2A Knockout Allergic Mice

S. Jamal Mustafa1, C. Ledent2, R. R. Morrison3 and A. Nadeem1

1Department of Physiology and Pharmacology and Center for Interdisciplinary Research in Cardiovascular Sciences, West Virginia University, Morgantown, West Virginia, USA. 2University of Brussels, Belgium; 3Division of Critical Care Medicine, St. Jude Children's Research Hospital, Memphis, Tennessee, USA.

Four types of adenosine receptors have been implicated in adenosine-mediated physiological and pharmacological responses in the lung. A2A receptor activation has been attributed to potent anti-inflammatory properties and its deficiency may lead to activation of pro-inflammatory transcription factor NF-kappa B. The present study was undertaken to investigate the role of adenosine A2A receptor on iNOS, Rel A (p65 subunit of NF-kappa B) and A2A adenosine receptor gene expression along with the levels of pro-inflammatory cytokines in the lung in an allergic asthma model [1, 2, 3, 4] in adenosine A2A receptor knockout (KO) mice. A2A KO and wild-type (WT) mice were sensitized with two i.p. injections of ragweed (200 µg) on days 1 and 6. Aerosolization challenge was performed with 0.5% ragweed for 20 minutes both in the morning and afternoon on days 11, 12 and 13. Twenty-four hrs after the last challenge, bronchoalveolar lavage fluid (BALF) and lungs were collected from the mice for cytokine and gene expression assays, respectively. Allergen challenge in sensitized mice increased gene expression of both Rel A and i-NOS of WT and KO as compared to their respective controls (p<0.01), with KO having greater expression of both of these genes than WT. A2A receptor expression was down-regulated by allergen challenge in WT sensitized mice as compared to WT controls whereas no transcripts of this receptor were detectable in KO mice. On the other hand, levels of pro-inflammatory cytokines (IL-2 and IL-4) were found to be higher in KO allergen challenged mice as compared to their respective WT sensitized and challenged mice (p<−0.01). Moreover, total nitrates and nitrites as a measure of nitric oxide production in the lung were also found to be higher in WT and KO allergen challenged mice as compared to their respective controls (p<0.01) with KO sensitized and challenged mice having greater NO production than respective WT mice (p<0.05). These data showed that A2A receptor down-regulation resulted in increased gene expression of Rel, A which is believed to be involved in transcription of pro-inflammatory genes. The corresponding increased levels of pro-inflammatory cytokines and expression of i-NOS along with increased NO generation in the lung may be a result of this Rel A activation. These data suggest that A2A receptor deficiency might result in increased expression of pro-inflammatory transcription factor Rel A leading to downstream expression of pro-inflammatory genes in this model of asthma. (Supported by HL-027339).

A2A Adenosine Receptor Antagonists: Synthesis, Biological Activities and Molecular Modelling Analysis of Polysubstituted Adenines

Diego Dal Ben,1 Gloria Cristalli,1 Catia Lambertucci,1 Sauro Vittori,1 Karl-Norbert Klotz,2 Rosaria Volpini1

1Dipartimento di Scienze Chimiche, Università di Camerino, via S. Agostino, 1, 62032 Camerino, Italy 2Department. of Pharmacology, University of Wuerzburg, Wuerzburg, Germany diego.dalben@unicam.it

The research on adenosine receptors has resulted in considerable progress in identifying selective agonists and antagonists and an increased understanding of the particular roles adenosine receptor subtypes play in physiological processes [1]. In the last years, A2A adenosine receptor antagonists have been proposed as an attractive pharmacological tool for the treatment of several diseases, such as motor dysfunctions, because of the interaction between A2A and D2 dopamine receptors in the basal ganglia, and A2A antagonists could improve motor dysfunction in a variety of models relevant to Parkinson_s disease. In addition, A2A adenosine receptor antagonists or the absence of A2A receptors is associated to neuroprotection. In this work our attention has been directed toward A2A adenosine receptor antagonists with adenine structure. In previous studies it was demonstrated that the introduction in 2 position of 9-ethyladenine of alkylamine or alkoxy chains leads to compounds endowed with affinity for the human adenosine receptors in the 2M range [2, 3]. In particular 9-ethyl-2-phenylethoxyadenine (2) and 9-ethyl-2-phenylethylaminoadenine (3) showing Ki A2A = 0.12 µM and 0.15 µM respectively, were slightly A2A selective. In addition, the introduction of a bromine atom in 8 position of 9-ethyladenine (1) led to 8-bromo-9-ethyladenine (4) endowed with improved affinity for all receptor subtypes. Starting from these observations and in the search for potent and selective adenosine receptor antagonists, a new series of 9-ethyladenine derivatives bearing different chains in 2 and 8 position was undertaken. Binding studies at human adenosine receptors showed that the new compounds are endowed with high affinity for the A2A receptor and are slightly selective for the same subtype. In particular, 8-bromo-9-ethyl-2-phenylethoxyadenine (5; Ki A2A = 0.002 µM) and 9-ethyl-8-furyl-2-phenylethoxyadenine (6; Ki A2A = 0.002 µM) resulted to be two of the most potent A2A antagonists with adenine structure reported so far. Molecular modelling studies have been performed by docking analysis of the compounds in a rhodopsin-based model of the human A2A receptor, and the results have been compared with the binding data. Open image in new window

Absence of NTPDase1 Unmasks Vasoconstrictor Effects of Extracellular Uracil Nucleotides in Mouse Aortic Rings

G. Kauffenstein1, S.C. Robson2, J. Sé vigny1.

1Centre de Recherche en Rhumatologie et Immunologie, Université Laval, Québec, Canada. 2Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA. Gilles.kauffenstein@crchul.ulaval.ca

Extracellular nucleotides regulate vascular tone in different ways and can promote both vasodilation and vasoconstriction. Activation of certain endothelial P2 receptors induces the release of endothelium derived vasodilating agents (NO, PGI2, EDHF) leading to a vasorelaxation. In contrast, the direct effect of nucleotides on vascular smooth muscle cells (SMC) P2 receptors induces the phosphorylation of myosin light chain, its association with actin leading to SMC contraction and a vasoconstriction. Nucleoside triphosphate diphosphohydrolase-1 (NTPDase1) is expressed on both vascular endothelium and smooth muscle. As NTPDase1 is the major ectonucleotidase regulating nucleotide concentrations and platelet activation in the environment of blood vessels, we propose that this enzyme may also regulate vasomotor tone. Despite normal adventitial expression of NTPDase2 in Entpd1 −/− mice vessels, the absence of NTPDase1 leads to a dramatic deficit in global nucleotidase activity (hydrolysis of ATP, ADP, UTP and UDP). Using ex vivo isometric contraction experiments we compared the vascular reactivity of aortic ring of Entpd1 −/− and +/+ mice. We found that the effect of exogenous nucleotides is enhanced in the absence of NTPDase1. While UDP and UTP (100 µM) induced a strong contraction (1.7 fold of Phenylephrine-induced contraction) in Entpd1 −/− mice aorta rings, only a weak contraction could be measured in Entpd1+/+ mice aorta rings (0.1 fold of Phenylephrine-induced contraction). The pharmacological profile of the response suggests that a P2Y6 like receptor is involved in the uracil nucleotide-induced vasoconstriction. Our data show that NTPDase1 regulates extracellular nucleotide signalling at the level of medial SMC.

Absence/Reduced Level of Diadenosine Triphosphatase (Fhit protein) is a Characteristic Feature of Several Neural Tumoural Cell Lines

P. Rotllán1, C.R. Rodríguez-Ferrer1, S. Oaknin1, D. Lorenzo-Villegas2 and E. Castro2

Departments of Biochemistry and Molecular Biology. Universities of 1La Laguna and 2Las Palmas de Gran Canaria. Canary Islands, Spain. protllan@ull.es

Mammalian cells contain several intracellular diadenosine polyphosphates, ApnA, a family of dinucleotides with intriguing biological functions. Among them, Ap3A and Ap4A receive most attention and appear to play antagonistic roles as intracellular signals in the control of cellular status. Cell proliferation and differentiation are associated with increases in Ap3A/Ap4A ratio but apoptosis with decreases. Increased intracellular Ap4A concentration may be involved in tumour cell growth suppression by inducing apoptosis. Diadenosine triphosphatase (Ap3Aase) and asymmetrical diadenosine tetraphosphatase (Ap4Aase) are the most important regulators of intracellular Ap3A and Ap4A levels. It has been reported that the novel tumour suppressor Fhit (fragil histidine triad) protein displays dinucleotide hydrolase in vitro and that the “old” enzyme Ap3Aase and Fhit are identical proteins. Re-expression of Fhit protein in Fhit deficient cancer cells results in growth inhibition and restoration of apoptosis.

We have determined the activities and expression of Fhit-Ap3Aase and Ap4Aase in soluble protein extracts from the following tumoural cell lines: C6 (rat glioma) Daoy (human medulloblastoma) and PC12 (rat pheochromocytoma). Soluble protein extracts from brain areas (hypothalamus, hippocampus, temporal cortex, frontal cortex and striatum), primary astrocytes from rat cerebellum and HCN-1A cells were used as controls. The fluorogenic substrates ε-(Ap3A) and ε-(Ap4A) were used for measurement of Ap3Aase and Ap4Aase activities and polyclonal Fhit antibodies to detect Fhit-Ap3Aase expression by western blotting. Both Fhit-Ap3Aase and Ap4Ase were clearly detected in all brain areas with activity levels around 4 and 22 mU/mg respectively and Ap4Aase/Ap3Aase activity ratios ranging between 4 and 6. Similar values were obtained for whole brain extracts. These ratios seem to be organ specific and limited between 4 and 10. Table I summarizes results obtained for cell cultures and whole brain. Table 1. Fhit-Ap3Aase and Ap4Aase activity (mU/mg protein) in extracts of whole brain and cell cultures. Values are means T SD. R means Ap4Aase/Ap3Aase ratio.
Table 1

Fhit-Ap3Aase and Ap4Aase activity (mU/mg protein) in extracts of whole brain and cell cultures. Values are means ± SD. R means Ap4Aase/Ap3Aase ratio.










4.22 ± 0.43

22.10 ± 2.21

5.23 ± 0.43


0.10 ± 0.03

5.46 ± 0.72

53.63 ± 8.19


2.25 ± 1.09

21.33 ± 5.42

10.05 ± 2.47


0.33 ± 0.53

15.98 ± 2.32

140.7 ± 163.8


0.40 ± 0.20

9.97 ± 3.88

39.64 ± 36.75



14.76 ± 3.76

Fhit-Ap3Aase activity was extremely low in PC12 and C6 cells and practically undetectable in Daoy cells. These tumoural cell lines presented aberrant high Ap4Aase/Ap3Aase ratios at expense of low Ap3Aase activity. Lack of Fhit-Ap3Aase activity correlated with lack of Fhit protein expression as detected by western-blotting. Results suggest that alterations in metabolism of ApnA and their intracellular antitumour signalling pathways are involved in genesis/progression of neural tumours.

Acknowledgments: Work funded by grants TR5-02 and TR2003/06 (Gobierno de Canarias with participation of Hospiten SA, NovoNordisk Ibé rica SA, Aventis Behring SA) and C03/06 (Red CIEN, FIS).

Activation of ecto-5′-nucleotidase by phospholipids

A. Pexa, S. Schudeja, A. Deussen

Department of Physiologie, Medical Faculty Carl Gustav Carus, TU Dresden, Dresden, Germany annette.pexa@tu-dresden.de


Ecto-5′-nucleotidase is the major enzyme controlling extracellular adenosine production in endothelial cells. Adenosine is a signal molecule involved in various hemodynamic and inflammatory processes. Activation of ecto-5′-nucleotidase increases adenosine concentration on the cell surface, and therefore may provide a tool to influence inflammatory processes. Furthermore, phospholipids are known modulators of various hemodynamic and inflammatory processes. Therefore adenosine might act as part of a signal transduction pathway in phospholipid signalling.


As experimental model we used human umbilical vein endothelial cells (HUVEC), pre-incubated with 5–10 µM of various phospholipids including γ-acyl-β-lyso-α-phosphatidylcholine, β-arachidonyl-γ-palmityl-α-phosphatidylcholine, β,γ-dipalmityl-α-phosphatidyl-choline, β,γ-dipalmityl-α-phosphatidylethanolamine, β,γ-dipalmityl-α-phosphatidylserine, γ-acyl-β-lyso-α-phosphatidylethanolamine, γ+-acyl-β-lyso-α-phosphatidylic acid, sphingosine-1-phosphate and sphingosylphosphorylcholine and β-acetyl-γ-O-hexadecyl-α-phosphatidylcholine (known as platelet activating factor). In the cell supernatant the extracellular dephosphorylation rate of the fluorescent AMP-analogue 1,N6-etheno-5′AMP to 1,N6-etheno-adenosine was measured by HPLC.


Out of these ten structurally related phospholipids only lysophosphatidylcholine (LPC), sphingosylphosphorylcholine (SPC) and platelet activating factor (PAF) increased the breakdown rate of AMP in a dose dependent manner. Interestingly, the activation can only be observed in a narrow concentration range between 5 and 20 µM. At higher phospholipid-concentrations the effect is blunted. Pharmacological blocking experiments with AOPCP, a specific inhibitor of ecto-5′-nucleotidase, showed that this effect was due to an activation of ecto-5′-nucleotidase.

Although protein kinases A and C are known to activate ecto-5′-nucleotidase, application of blockers of both kinases did not diminish the effects of LPC, SPC and PAF. Furthermore, the effect does not seem to be mediated by G-protein-coupled receptors, as suramine and pertussis toxin had no effect.


Using information on the known molecular structures of tested phospholipids, a phosphatidylcholine residue in α-position and a short chain length fatty acid esterified in β-position seem essential for activation of ecto-5′-nucleotidase by phospholipids. The signal transduction/activation mechanism of ecto-5′-nucleotidase remains unclear at present.

Activation of P2Y1 and P2X7 receptors induce calcium/calmodulin-dependent protein kinase II phosphorylation in cerebellar granule neurons.

MaTeresa Miras-Portugal(1), David León (1), Patricia Marín Garcia(1), Felipe Ortega(1), Jesú s Sánchez-Nogueiro(1), Cristina Hervás (1), Nyrup M (2)

(1) Department of Biochemistry, Veterinary Faculty, Universidad Complutense de Madrid. (2) Department of Pharmacology and Pharmacotherapy. The Danish University of Pharmaceutical Science. mtmiras@vet.ucm.es

The activation of nucleotide receptors -both ionotropic, P2X, and most of metabotropic, P2Y- increases intracellular calcium concentration, resulting in calcium-calmodulin-dependent protein kinase II (CaMKII) activation. Stimulation of cerebellar granule neurons in culture -with different P2X and P2Y agonists and their effect on CaMKII phosphorylation- was studied using immunocytochemical and microfluorimetrical techniques. P2X agonist: 2′–3′-o-(4-benzoylbenzoyl)-adenosine 5′-triphosphate (BzATP), α,β-methylene adenosine 5′-triphosphate (α,β-meATP) and diadenosine pentaphosphate (Ap5A), and P2Y agonists: 2-(methylthyo)-adenosine diphosphate (2MeSADP) and uridine 5′-bisphosphate (UDP), tested induced a CaMKII phosphorylation but with a different immunostaining pattern in each group. Stimulation with 2MeSADP induced a Ca2+ release from intracellular stores and a significant CaMKII phosphorylation in both cell somas and fibres. This agrees with the subcellular distribution of P2Y1. MRS 2179, a specific P2Y1 inhibitor, antagonized the 2MeSADP effect. On the other hand, cerebellar granule neuron stimulation with BzATP, in Mg2+ free conditions, produced extracellular calcium entrance and, as a result, a significant increase in CaMKII phosphorylation mostly in fibers, which correspond with P2X7 subdistribution. Immunocytochemical and microfluorimetrical experiments, using Zn2+ and Brilliant Blue G (BBG), as an specific P2X7 antagonist, confirmed that BzATP was acting through the P2X7 receptor. These results, indicate that P2Y1 and P2X7, produce a significant increase in CaMKII phosphorylation, but show important differences in subcellular distribution and in effect duration. The abundant presence of P2X7 at the synaptic structures suggest the relevant role played by this receptor in synaptic plasticity.

Activation of the adenosine A2A receptor protects mice deficient in CD8+ T-lymphocytes from liver ischemia-reperfusion injury.

McGreevy KS, Linden J.

Department of Medicine, University of Virginia, Charlottesville, VA 22908 USA ksm3@hscmail.mcc.virginia.edu

BACKGROUND: The actions of T-lymphocyte subsets and their regulation through adenosine is of interest in many disease models, including ischemia-reperfusion injury. This study explores the effect of the selective adenosine A2A receptor agonist ATL-313 in animals depleted of CD4+ or CD8+ T-lymphocytes in a model of liver ischemia-reperfusion. METHODS: Immune-competent B6 mice were depleted of CD4+ or CD8+ T-lymphocytes in vivo by the introduction of an antibody to the respective cell type into the circulation. Rat monoclonal antibodies were prepared by an intraperitoneal injection of the hybridoma cell lines GK1.5 or 2.43 into SCID mice. Ascites fluid was collected and purified, and the final antibody concentration was adjusted to 1 mg/ml for storage. Experimental animals were tested for appropriate depletion of CD4+ or CD8+ T-lymphocytes through FACS analysis of the remaining T-lymphocyte population, labeled with commercially available antibodies. The antibodies were titered to determine the optimal dose of each for producing effective and selective depletion, found to be 0.05 mg per mouse delivered on two consecutive days. On the fourth day after the second injection animals were subjected to liver ischemia/reperfusion, which was performed by application of a microaneurysm clamp to the hepatic triad for seventy-five minutes. Drug-treated animals received a bolus injection of 3 µg/kg ATL-313 fifteen minutes prior to reperfusion, and an Alzet pump delivering 1 ng/kg/min ATL-313 at reperfusion, while control animals received vehicle. Blood was drawn 24 hours after the onset of ischemia, and a serum alanine aminotransferase (ALT) assay was performed to assess the degree of liver injury. RESULTS: In vehicle treated animals, the extent of injury in CD4+ depleted mice (1846.2 ± 440.8 IU/L) was significantly less than that of wildtype (5747.9 ± 413.6 IU/L) or CD8+ depleted mice (5809.4 ± 1097.1 IU/L). ATL-313 provided significant protection in wild-type (3472.0 ± 666.0 IU/L) and CD8+ depleted animals (3475.5 ± 424.9 IU/L) but not in CD4+ depleted animals (1768.9 ± 735.2 IU/L). CONCLUSION: The loss of CD4+ T-lymphocytes provides protection from liver ischemia-reperfusion injury, while the loss of CD8+ cells does not. A protective effect of the adenosine A2A receptor agonist ATL-313 is seen in wild-type and CD8+ depleted animals, but no further protection is conferred upon CD4+ depleted animals, suggesting that ATL-313 protects through an interaction with CD4+ cells.

Activation-dependent Trafficking of NTPDase2 in Chinese Hamster Ovary Cells

SM Vlajkovic1, CJH Wang1, C. Soeller1, N. Braun3, H. Zimmermann3, PR Thorne2, GD Housley1

1Department of Physiology and 2Discipline of Audiology, Faculty of Medical and Health Sciences, The University of Auckland, New Zealand; 3Biozentrum der J.W. Goethe-Universitä t, AK Neurochemie, Frankfurt am Main, Germany s.vlajkovic@auckland.ac.nz

Membrane-bound NTPDase2 is a member of the ecto-nucleoside triphosphate diphosphohydrolase (E-NTPDase) enzyme family involved in the regulation of P2 receptor signalling. NTPDase2 has broad substrate specificity for extracellular nucleotides, but hydrolyses nucleoside 5′-triphosphates with high preference over nucleoside 5′-diphosphates. In this study we have sought to determine how enzyme substrates acting on P2 receptors affect intracellular NTPDase2 trafficking. To achieve this, Chinese hamster ovary (CHO) cells expressing endogenous P2X7 and P2Y2 receptor subunits were transiently transfected with rat-specific NTPDase2 cDNA (GenBank accession no. NM_172030) tagged with a green fluorescent protein (GFP) to allow direct visualisation of subcellular localisation and trafficking of NTPDase2.


Cells were transfected with plasmid containing recombinant GFP-NTPDase2 cDNA using Lipofectamine™ 2000 reagent. Transiently transfected CHO cells were imaged using an inverted confocal microscope (Zeiss LSM410) with 488-nm excitation. Cells were perfused (1 ml/min) with NTPDase2 substrates (ATP, UTP) and non-hydrolyzable ATP and ADP analogues ATPγS and ADPβS (0.5 mM) in phosphate-free saline solution at room temperature. Z-stack image series were taken at 0, 5, 10, 20 and 30 minutes. P2 receptor inhibitors (suramin, PPADS, brilliant blue) were pre-incubated with CHO cells for 60 minutes before adding extracellular nucleotides. NTPDase2 incorporation into the cell membrane was determined at time intervals by comparative analysis of pixel intensity of fluorescence in the cytosolic and membrane compartments. To delineate the plasma membrane, CHO cells were stained with a membrane dye Di 4 ANNEPS (Molecular Probes). For calcium imaging, cells were loaded with 4 µM Fluo-4 (Molecular Probes) in cell culture medium for 60 minutes before stimulation with ATP, UTP and ADP. This study confirmed the functional expression of endogenous ATP-preferring P2X receptors and ADP and UTP-preferring P2Y receptors. NTPDase2 activity was measured in transiently transfected GFP-NTPDase2 CHO cells. Cells were perfused with NTPDase2 substrates ATP and UTP (0.5 mM) individually to assess the hydrolysis rate. Perfusion was paused for 1 minute at time intervals used for cell imaging and a sample aliquot was removed for HPLC analysis.

Results and conclusion

(1) The present study shows that NTPDase2 tagged with GFP is functional. (2) ATP and its non-hydrolysable analogue ATPγS induce rapid membrane incorporation of NTPDase2 from putative intracellular stores in CHO cells transiently transfected with GFP-NTPDase2, whilst UTP and ADPβS are ineffective. (3) NTPDase2 trafficking requires extracellular Ca2+. (4) Pharmacological studies show that activation of P2X receptors involving Ca2+ entry, rather than P2Y receptors releasing stored Ca2+, may be linked to trafficking of NTPDase2 to the cell membrane. (5) Initial hydrolysis rate for ATP correlates with increased NTPDase2 expression on the plasma membrane, whilst the initial velocity of UTP hydrolysis remains steady. (6) Increased trafficking of NTPDase2 to the cell membrane may reflect a regulatory mechanism to limit excessive stimulation and desensitisation of P2 receptors.

Supported by the Health Research Council of New Zealand and Auckland Medical Research Foundation.

Acute elevation of free fatty acids induces vasodilation which is not mediated by adenosine receptor stimulation

NP Riksen1,2, M Bosselaar2, SJL Bakker3, RJ Heine4, GA Rongen1,2, CJ Tack2, P Smits1,2

1Dept of Pharmacology-Toxicology, and 2Internal Medicine, Radboud University Nijmegen Medical Centre, 3Department of Internal Medicine, University Medical Centre Groningen, 4Diabetes Centre, VU University Medical Center, the Netherlands. N.Riksen@aig.umcn.nl


Characteristics of the metabolic syndrome are associated with a hyperdynamic circulation. In these patients, a hyperdynamic circulation predicts future development of type 2 diabetes mellitus. Free fatty acids (FFA) have been implicated as mediators of this hyperdynamic circulation in patients with the metabolic syndrome. Previous in vitro studies have shown that FFA can inhibit the mitochondrial adenine nucleotide transporter, thus increasing cytosolic adenosine concentration. We hypothesized that FFA-induced vasodilation is mediated by adenosine receptor stimulation1.


Nine healthy subjects participated in a randomised cross-over trial. Intralipid was administered intravenously for 2 hours with heparin to activate lipoprotein lipase to elevate plasma FFA. Glycerol/heparin was given on the other occasion as Control treatment. Forearm blood flow (FBF) was measured by venous occlusion plethysmography. During the last 15 minutes of infusion, the adenosine receptor antagonist caffeine was administered into the brachial artery of the non-dominant arm. In a validation study, adenosine was infused into the brachial artery (5 µg/min per dl of forearm tissue) and subsequently caffeine was added (90 µg/min/dl) to confirm effective adenosine receptor antagonism (n = 6).


Administration of Intralipid increased plasma FFA from 334 ± 108 µmol/l at baseline to 1269 ± 234, whereas Control infusion increased FFA from 338 ± 114 µmol/l to only 432 ± 175. Intralipid infusion induced a hyperdynamic circulation: systolic blood pressure increased from 115 ± 8 to 122 ± 8 and from 116 ± 10 to 120 ± 9 mmHg, pulse pressure increased from 53 ± 6 to 59 ± 6 and from 53 ± 5 to 55 ± 9 mmHg and FBF increased from 1.8 ± 0.7 to 2.7 ± 1.8 and from 2.3 ± 0.7 to 2.7 ± 1.8 ml/dl/min for Intralipid versus Control infusion, respectively. (P = 0.05, P < 0.05, and P < 0.05). In the validation study, adenosine-induced vasodilation was effectively reduced by caffeine (P < 0.01). However, the FFA-induced vasodilation could not be reduced by administration of caffeine (P = NS).


Acute elevation of plasma FFA levels induces an increased skeletal muscle blood flow, systolic blood pressure, and pulse pressure (hyperdynamic circulation). The vasodilator response is not mediated by adenosine receptor stimulation. Therefore, it is unlikely that adenosine mediates the association of a hyperdynamic circulation and the metabolic syndrome.

Acyl derivatives of coenzyme A inhibit ADP-induced platelet function via antagonism at P2Y1 and P2Y12 receptors.

Stan Heptinstall, Panagiotis Manolopoulos, Jacqueline R Glenn, Susan C Fox, Jane A May, Natalia Dovlatova, Siau-Wei Tang, Nadine Bauer, Martin Hesse, Neil R Thomas1 and Vera Ralevic.

Centre for Integrated Systems Biology and Medicine and 1Centre for Biomolecular Sciences, University of Nottingham, Nottingham, UK. s.heptinstall@nottingham.ac.uk

ADP plays a central role in platelet aggregation and is important in haemostasis and thrombosis. Individuals whose platelets respond poorly to ADP often present with easy bruising or bleeding. Also a pharmaceutical agent that inhibits ADP-induced platelet aggregation, clopidogrel, is in widespread use as an antithrombotic agent. ADP induces platelet aggregation acting via P2Y1 and P2Y12 receptors and clopidogrel acts as an antagonist at P2Y12 receptors. Interaction with P2Y1 induces rapid Ca2+ mobilisation leading to a change in the shape of platelets from smooth discs to spiny spheres and a transient aggregation response. Interaction with P2Y12 inhibits adenylate cyclase and activates PI-3-kinase which sustains the aggregation response. Both receptor subtypes are required for full aggregation induced by ADP. Antagonists acting at either P2Y1 or P2Y12 receptors (or both) could be usefully employed as anti-thrombotic agents. Following an observation that derivatives of coenzyme A (CoA) act as antagonists at P2Y1 receptors transfected into oocytes (1), we tested the effects of various acyl derivatives of CoA on platelet aggregation. Measurements were performed in both whole blood and platelet-rich plasma. CoAs containing a range of saturated and unsaturated acyl groups inhibited ADP-induced platelet aggregation. Those with saturated acyl groups containing 16–18 carbons were most effective and introduction of unsaturation into the fatty acid chains reduced efficacy. The effects of palmitoyl-CoA (16:0) were then tested on a variety of aspects of platelet function. Specificity for effects on platelet aggregation induced by ADP was demonstrated by showing that palmitoyl-CoA inhibited ADP-induced aggregation but not that induced by platelet activating factor, thrombin receptor activating peptide (TRAP) or serotonin. ADP-induced aggregation and also P-selectin expression (a marker of α-granule release) were inhibited in a competitive way. Measurements of shape change were performed using the Biola laser aggregometer and also flow cytometry. Complete inhibition of ADPinduced shape change by palmitoyl-CoA indicated antagonism via P2Y1 receptors. Again there was no inhibition when other agonists were used, including agonists that act at P2X1 receptors. Ca2+ mobilisation was measured using fluo 3-labelled platelets and flow cytometry. These studies also revealed complete inhibition by palmitoyl CoA of the increase in [Ca2+]i that occurred in response to ADP, again indicative of antagonism at P2Y1 receptors. Experiments were also performed to look for possible antagonistic effects of palmitoyl-CoA at P2Y12 receptors on platelets. This was done by examining platelet aggregation and/or Ca2+ mobilisation in response to collagen and to pairs of agonists: ADP and serotonin, and ADP and TRAP. Measurements of ADP-induced changes in cAMP and vasodilator-stimulated phosphoprotein phosphorylation were also performed. Results obtained with palmitoyl-CoA were compared with those for MRS 2179 (a P2Y1 antagonist), AR-C69931 (a P2Y12 antagonist) and combinations of these. The results were all consistent with palmitoyl-CoA inhibiting platelet function via antagonism at P2Y12 receptors in addition to P2Y1 receptors. Acyl derivatives of CoA such as those used in this investigation are natural agents that could be released into blood following cell and tissue activation and/or damage and thereby be involved in endogenous control of platelet function. They might also form the basis for a new class of antithrombotic drugs.

Adenine and adenosine salvage pathways in erythrocytes and the role of S-adenosylhomocysteine hydrolase. A theoretical study using elementary flux modes

Dimitar Kenanov and Stefan Schuster

Department of Bioinformatics, Friedrich Schiller University, Ernst-Abbe-Platz 2, 07743 Jena, Germany Phone +49-3641-949580, Fax +49-3641-946452 kenanov@minet.uni-jena.de

This talk is devoted to the study of redundancy and yield of salvage pathways in human erythrocytes. These cells are well known not to be able to synthesize ATP de novo. However, the salvage (recycling) of certain nucleosides or bases to give nucleotide triphosphates is operative. As the salvage pathways use enzymes consuming ATP as well as enzymes producing ATP, it is not easy to observe the net synthesis of ATP. As for pathways using adenosine, a straightforward assumption is that these pathways start with adenosine kinase. However, a pathway bypassing this enzyme and using S-adenosylhomocysteine hydrolase instead was reported. So far, this route has not been analysed in detail.

For the analysis we use the concept of elementary flux modes which can be described as a minimal set of enzymes able to operate at steady state with all irreversible reactions running in the right direction. This method opens up the possibility to study the various modes of behaviour of the biochemical system and it also allows detection of possible bypasses otherwise not easily understood by just observing the metabolic pathway. This tool enables one to simulate the interaction between several subsystems utilising substrates of interest or systems with enzyme deficiencies which can be used in investigating diseases caused be these deficiencies.

Using the concept of elementary flux modes, in the present study we investigate theoretically which salvage pathways exist in erythrocytes, which enzymes belong to each of these and what relative fluxes these enzymes carry. Here, we compute the net overall stoichiometry of ATP build-up from the recycled substrates and show that the network has considerable redundancy. For example, four different pathways of adenine salvage and 12 different pathways of adenosine salvage are obtained. They give different ATP/glucose yields, the highest being 3 : 10 for adenine salvage and 2: 3 for adenosine salvage provided that adenosine is not used as an energy source. Implications for deficiencies of the enzymes ADA and PNPase are also discussed.

Adenosine 5′-triphosphate (ATP) in exhaled breath condensate of patients with hypoxia

Zs Laázaár1, É Huszár1, M Valyon2, I Horváth1

1Department of Pathophysiology, 2Department of Clinical Diagnostics, National Korányi Institute of TB and Pulmonology, Budapest, Hungary zsofia.lazar@koranyi.hu

ATP is released from many different cells in response to hypoxia activating purine receptors expressed by the same or neighboring cells leading to bronchoconstriction in the airways. We hypothesized that in hypoxia ATP concentration in the lungs is elevated compared to normal oxygenation. In our study we compared ATP concentrations in exhaled breath condensate (EBC) from patients with respiratory diseases causing hypoxia and healthy subjects, we studied whether clinical improvement of patients resulted in a decrease in ATP concentrations and investigated the correlation between ATP levels and partial arterial oxygen pressure. EBC was collected for 15 minutes from 15 patients with hypoxia and 8 healthy subjects. Simultaneously, blood gas of all participants was analyzed. In 9 subjects, EBC collection and blood gas analysis were repeated following clinical improvement. ATP was measured by a bioluminescence assay (ENLITEN ATP Assay System, Promega). Statistical analysis was performed non-parametrically. Data are expressed as median [range]. Patients exhibited profound hypoxia compared to healthy subjects (7.51 [4.67–8.62] kPa vs. 10.23 [8.32–13.47] kPa, p < 0.01) but there was no significant difference in EBC ATP concentration between the two groups (40 [22–473] pM vs. 33 [26–257] pM). Although clinical variables and pO2 demonstrate significant improvement during hospitalization (n = 9, pO2 pretreated: 7.84 [5.07–8.62] kPa vs. pO2 at discharge: 8.85 [5.08–9.75] kPa, p < 0.01), ATP values did not change significantly (n = 9, ATP pretreated: 26 [22–473] pM vs. ATP at discharge: 47 [18–77] pM) and there was no correlation between changes in pO2 and ATP concentration in EBC. In summary, ATP is detectable in EBC; however, our results suggest that hypoxia per se is not a major contributing factor to its level.

This study was supported by the Hungarian National Scientific Research Foundation (OTKA T-43396).

Adenosine A1 and A2A Receptors Regulate Response to Infection with Schistosomiasis (S.Mansoni)

Tuere F. Wilder1, Mazen Makarem2, Stephen J. Davies2 and Bruce N. Cronstein1

1Division of Clinical Pharmacology, Department of Medicine, NYU School of Medicine, New York, NY USA.2

Department of Microbiology and Immunology, USUHS, Bethesda, MD USA Tuere.wilder@med.nyu.edu


Schistosomiasis is a parasitic disease affecting more than 200 million people worldwide in which the pathology, including hepatic fibrosis and portal hypertension, results from host reactions to parasite. We have documented that adenosine released endogenously ligates adenosine A2A receptors to promote fibrosis but inhibit inflammation. It also has been documented that A1 receptor activation may serve to regulate pulmonary and renal inflammation in mice during chronic diseases. We therefore determined whether adenosine A1 and A2A receptors contribute to the pathophysiology of schistosomal infection.


3–10 month old female A1 and A2A KO and WT littermate controls were infected with S. mansoni and sacrificed 4 or 8 weeks post-infection (p.i.). Worms were perfused from the portal vein of infected mice and measured. Fibrosis was digitally evaluated in 6 picrosirius red-stained sections/mouse. mRNA for IFN-γ, IL-4, and IL-10 from whole livers was analyzed by real time RT-PCR. Supernates of cultured mesenteric lymph node (MLN) cells stimulated with soluble egg antigen were assayed for IFN-γ, IL-4, IL-10 and IL-5 by ELISA.


The worms from the A1KO mice were larger than their WT littermates 4 and 8 weeks p.i. There was also an increase in the percentage of females in pairs in the A1KO in comparison to the WT (76 ± 6 N = 8 vs. 36 ± 19 N = 3, p = 0.02) 4 weeks p.i. In the A2AKO mice there was a 5 and 20-fold increase in IFN-γ and IL-4 message, respectively, 4 weeks p.i. This increase was accompanied by a striking 2-fold increase in hepatic fibrosis in the A2AKO compared to the WT mice (19 ± 2 N = 8 vs. 8 ± 1 N = 7, p < 0.0001). In addition, the levels of all cytokines tested in supernates from cultured MLN cells showed at least a 2-fold increase in the A1KO and A2AKO mice as compared to their WT controls. There were even more marked increases in IFN-+ and IL-4 levels in the KO mice (Table I).

A1WT ng/ml)

A1KO (ng/ml)

P value

Increase between groups

A2AWT (ng/ml)

A2AKO (ng/ml)

P value

Increase between groups


170 ± 7 N = 3

877 ± 9 N = 3



152 ± 7 N = 3

1409 ± 17 N = 3





31 ± 1 N = 3

140 ± 3 N = 3




Endogenous adenosine modulates immune responses to schistosomal infection via interaction with its receptors. Increased hepatic fibrosis in infected A2A KO mice most likely reflects enhanced immune reactions to parasites observed in these animals. On the other hand, although A1 receptors modulate the immune response they play no role in the development of hepatic fibrosis in infected mice but appear to play a permissive role in worm fecundity.

Adenosine A1 and A2A Receptors, CD39/CD73KO Influence Granulomatous Responses to Mycobacterium Bovis BCG

Firas M. Kara1, M. Carmen Montesinos2, Adam Mor1, Simon Robson3, Linda Thompson4 and Bruce N. Cronstein1

1Division of Clinical Pharmacology, Department of Medicine, NYU School of Medicine, New York, NY USA,

2University of Valencia, Valencia, Span, 3Harvard School of Medicine, Boston, MA. 4Oklahoma Medical Research Foundation, Oklahoma City, OK Firas.kara@med.nyu.edu


Granuloma formation around infected macrophage is a defining cellular response to mycobacterium infections. Granulomas eliminate bacteria and also protect surrounding host tissue from destructive inflammatory responses. Without granuloma formation, mycobacterial infections can become widely disseminated and frequently lethal, as occurs in human AIDS-associated tuberculosis. Granuloma formation is characterized by a nodular aggregation of mononuclear cells and multinucleated giant cells. Adenosine, formed extracellularly from adenine nucleotides by nucleoside triphosphate dephosphorylase (CD39) and ecto-5′nucleotidase (CD73), modulates macrophage function by engaging cell surface adenosine receptors: A1 adenosine receptor activation promotes invitro giant cell formation and enhances phagocytosis, whereas A2A receptor activation inhibits in vitro giant cell formation and phagocytosis. Therefore, we sought to study the formation of hepatic granulomas induced by BCG in receptor (A1 and A2A) and adenosine-producing enzyme (CD39), (CD73) and (CD39/CD73) knockout mice.


6 week old CD39KO, CD73KO and CD39/CD73KO, A2A KO and A1KO and appropriate WT female mice were injected (IP) with BCG (107 CFU). After 4 weeks mice were sacrificed and hepatic granuloma number, morphology and size analyzed digitally and expressed as granulomas/10× field and pixels/granuloma, respectively.


Granuloma size did not differ among CD39KO, CD73KO and WT mice (3445 ± 796, N = 3, 2818 ± 610, N = 2 vs. 2767 ± 205, N = 10) but were larger in CD39/73KO mice (8295 T 2681, N = 5, P < 0.0001 vs. WT mice). There were similar numbers of granulomas in the CD73KO and CD39/73KO mice and WT mice (47 ± 14 N = 2 and 35 ± 12 N = 5 vs. 53 ± 5 N = 10, respectively). CD39KO mice showed increased number of significantly granulomas than wild type mice (63 ± 26 N = 3 vs. 53 ± 5 N = 10, P = 0.02).

A1KO mice formed more granulomas than WT mice (22 ± 9 vs. 13 ± 6, n = 5 for both, p = 0.004) although granulomas were smaller in A1KO mice (1135 ± 102 vs. 1611 ± 136, n = 5 for both, p = 0.004). In contrast, there were fewer granulomas in the A2AKO mice formed fewer granulomas (4 ± 2 vs. 12 ± 6, n = 5 for both, p = 0.00005) of the same size as WTs (Data not shown).


Adenosine, produced extracellularly from adenine nucleotides, alters the response to mycobacterial infections via interaction with A1 and A2A receptors on macrophages.

Adenosine A1 receptor occupancy and sleep regulation: a positron emission tomography study

Geissler E1,2, Ametamey SM1, Wyss M3, Treyer V3, Rétey JV2, Adam M2, Achermann P2, Schubiger PA1, {urLandolt HP}2

1Center for Radiopharmaceutical Science of ETH, PSI and USZ, ETH Zürich, Switzerland

2Institute of Pharmacology and Toxicology, University of Zürich, Switzerland

3Division of Nuclear Medicine, University Hospital Z:rich, Switzerland landolt@pharma.unizh.ch

Recent findings support the notion that adenosine and adenosine receptors play an important role in human sleep regulation1,2. Post-mortem studies demonstrated a high density of adenosine A1 receptors (A1AR) in brain regions involved in sleep regulation, such as thalamus, hippocampus, striatum, basal ganglia and cerebral cortex. Moreover, the non selective adenosine A1 and A2A receptor antagonist, caffeine, stimulates alertness and attenuates changes in the waking and sleep EEG, which are typically found after sleep deprivation. Nevertheless, the distinct roles of A1 and A2A receptors for sleep regulation are still controversial. The recent development of the potent and selective A1AR antagonist, 8-cyclopentyl-3-(3-18F-fluoropropyl)-1-propyl-xanthine (18F-CPFPX), offers the opportunity to directly visualize and quantify A1AR binding in the living human brain.

We established the radiosynthesis of 18F-CPFPX according to a recently reported two step reaction sequence3. We obtained the target compound 18F-CPFPX in a radiochemical yield of about 10% and a radiochemical purity of >99%. Specific activity exceeded 100 GBq/µmol.

In a still ongoing study, we combine quantitative EEG recordings and positron emission tomography (PET) brain imaging to determine whether A1AR occupancy is enhanced after 32 hours of prolonged wakefulness when compared to a baseline recording (after 8 hours of wakefulness). Moreover, we examine whether the sleep-deprivation induced changes in subjective vigilance, neurobehavioral performance and the EEG are reduced after 300 mg slowrelease caffeine intake when compared to placebo. To date, two groups of 5 healthy young men completed the study. The first results support a high 18F-CPFPX accumulation in sleep regulatory areas such as thalamus and cortex. In contrast, low ligand binding is found in cerebellum and brain stem.

We conclude that PET brain imaging with 18F-CPFPX is a promising new tool to gain new insights into adenosinergic mechanisms of sleep regulation in humans. The effects of sleep deprivation and caffeine on cortical and subcortical A1AR occupancy remain to be determined.

Research supported by a Center for Neuroscience Zürich PhD Grant and the Swiss National Science Foundation.

Adenosine A1 receptors are increased and sensitized in pick disease frontal cortex

José Luis Albasanz1, Agustín Rodriguez2, Isidro Ferrer2,3 and Mairena Martín1

1Departamento de Química Inorgánica, Orgánica y Bioquímica, Facultad de Químicas, Centro Regional de Investigaciones Biomédicas, Universidad de Castilla-La Mancha, Ciudad Real.

2Departamento de Biología Celular y Anatomía Patológica, Facultad de Medicina, Universidad de Barcelona, campus de Bellvitge, Hospitalet de Llobregat.

3Instituto de Neuropatología, Servicio de Anatomía Patológica, IDIBELL-Hospital Universitario de Bellvitge, Hospitalet de Llobregat. SPAIN jose.albasanz@uclm.es

Adenosine is a nucleoside widely distributed in nervous system where acts as a neuromodulator and neuroprotector through specific receptors. Adenosine receptors have been classified into four types: A1, A2A, A2B and A3 receptors. A1 and A3 receptors inhibit adenylyl cyclase activity through Gi/o proteins, while A2A and A2B receptors stimulate adenylyl cyclase through Gs proteins. Adenosine acting through A1 receptors inhibits excitatory neurotransmitter release, therefore acting as a neuroprotector. Pick's disease (PiD) is a fronto-temporal dementia characterized by severe atrophy of the frontal and temporal lobes that spares the pre-central gyrus and the posterior two-thirds of the superior temporal gyrus. This is accompanied by marked neuron loss, mainly in the upper cortical layers, and the appearance of typical phospho-tau-immunoreactive intraneuronal inclusions named Pick bodies, principally in the dentate gyrus of the hippocampus, CA1 region of the hippocampus, amygdala, septal nuclei, and upper layers of the entorhinal cortex and isocortex, together with phospho-tau-immunoreactive thorn-shaped and ramified astrocytes, and tau-positive bodies in oligodendroglia. The aim of the present work was to study A1 receptors/adenylyl cyclase (AC) pathway in post-mortem human cortex from Pick's disease (PiD) as compared with age matched non-demented controls. Adenosine A1 receptors, determined by radioligand binding and Western-blotting assays were significantly increased in PiD samples, suggesting the up-regulation of this receptor. Real time PCR analysis revealed an increase in mRNA coding A1 receptor in PiD cases. Adenylyl cyclase activity was determined in basal and stimulated conditions via stimulatory guanine nucleotide binding proteins (Gs) using GTP, or directly with forskolin. Basal AC activity was reduced in brains from PiD cases. This agrees with the decrease in AC I level detected by Western-blotting. No significant differences in GTP- or forskolin-stimulated adenylyl cyclase activity were observed between PiD and control cases. However, inhibition of forskolin-stimulated AC activity by a selective A1R agonist was significantly increased in brains from PiD. These results show that adenosine A1 receptors/adenylyl cyclase transduction pathway is up-regulated and sensitized in cortex brain from Pick's disease.

Adenosine A1 receptors in the phrenic motoneurons of adult rats: A biochemical study

Saharan SR, Kizy T, and Nantwi, KD.

Department of Anatomy and Cell Biology, Wayne State University, School of Medicine, 540 E. Canfield, Detroit, MI 48202 knantwi@med.wayne.edu

Previous studies from our laboratory have demonstrated that in an animal model of spinal cord injury (SCI), a latent respiratory motor pathway can be activated to restore respiratory activity to a hemidiaphragm paralyzed by an upper cervical (C2) spinal cord hemisection. Activation of the motor pathway occurs during a reflex known as the “crossed phrenic phenomenon”, by asphyxia or systemically administered theophylline. Theophylline acts via blockade of adenosine A1 receptors. Adenosine receptors are located in the phrenic motor neurons (PMN). The objective of the present study is to characterize adenosine receptors in the PMN of adult rats by radioligand binding. Binding Assays: The specific binding of the adenosine A1- receptor selective ligand [3H]-DPCPX to spinal cord membranes was determined using a rapid filtration assay. Spinal cord membranes were first incubated with adenosine deaminase (ADA, 5 U/mg protein) in 50 mM Tris buffer, 2 mM MgCl2 at pH 7.4 to remove endogenous adenosine. Membrane suspension (at 0.1 mg/ml protein in a final volume of 500 µl) was incubated for 1 h at 4-C with [3H]-DPCPX in the absence of a competing compound and equilibrium binding reactions were terminated by vacuum filtration (Brandel harvester GF/B filters 0.5%PEI). Each filter was washed four times with 4 ml of cold 50 mM Tris-HCl, pH7.4. Filter discs were punched into vials using a Deposit/Dispenser. Three milliliter aliquots of ‘Ultima Gold” Scintillation Cocktail were added into each vial prior to scintillation counting. Filter-bound radioactivity was determined by liquid scintillation counting using a Tricarb-2800-TR Liquid Scintillation counter. [3H]-DPCPX was used (at 8–12 concentrations) in saturation binding. Competition studies used theophylline as the competing ligand at ten concentrations (20 mM–20 pM); against a standard concentration of [3H]-DPCPX (1 nM). Binding data were analyzed with (Graphpad software,). Results: In naïve animals, a single receptor binding site with Bmax and Kd values of 256.00 ± 32.13 (fmol/mg protein) and 2.89 ± 0.45 (nM), respectively, was detected. Theophylline induced detection of a second binding site with a Bmax of 492.6 ± 3.15 (fmol/mg) and Kd 14.09 ± 2.06 (nM) respectively. Hemisection induced an up-regulation of A1 receptors; a Bmax 316.6 ± 25.10 fmol/mg protein without apparent change in affinity (Kd 2.72 ± 0.72 nM). It is concluded that A1 receptors in the PMN of adult rats are altered significantly by theophylline administration and C2 hemisection. The changes may underlie respiratory effects of theophylline. Furthermore, the up-regulation of the receptors after hemisection may be involved in theophylline-induced functional recovery.

Adenosine A1 Receptors Play a Critical Role in Osteoclast Formation

Firas M. Kara1, Bertil Fredholm2 and Bruce N. Cronstein1

1Division of Clinical Pharmacology, Department of Medicine, NYU School of Medicine, New York, NY USA.

2Karolinska Institutet, Stockholm, Sweden Firas.kara@med.nyu.edu


Osteoclasts are bone-resorbing, multinucleated giant cells that are essential for bone remodeling that are formed through the fusion of mononuclear precursor cells. Osteoclasts differentiate from hemopoietic precursors of the monocyte/macrophage lineage in the presence of M-CSF and receptor activator of NF-κB ligand (RANKL). Deficiency of osteoclasts leads to osteopetrosis, a condition characterized by increased bone density. Nonetheless, most bone diseases are due to increased bone resorption by osteoclasts and inhibition of osteoclastmediated bone resorption is a primary therapeutic objective. Indeed, most current therapies for osteoporosis are directed at inhibition of osteoclast function. Because we have previously reported that adenosine A1 receptor occupancy is required for fusion of stimulated human monocytes to form giant cells in vitro we determined whether there was a similar requirement for A1 receptor occupancy in osteoclast formation.


Spleens were harvested from female mice and, following isolation, splenocytes were resuspended in αMEM containing 10% FBS. The cells were cultured in the presence of M-CSF (30 ng/ml) and RANKL (30 ng/ml) with or without various concentrations of the A1 receptor antagonist DPCPX or the A1 receptor agonist N6- CPA. The culture was fed every 3 days by replacing half of the media containing the M-SCF, RANKL and adenosine receptor agonist/antagonist. After incubation for 7 days, cells were stained for tartrate-resistant acid phosphatase (TRAP). Osteoclasts were identified as TRAP-positive cells with 3 or more nuclei; the number of TRAP-positive multinucleated cells/well was then enumerated.


We found that adenosine A1 receptor occupancy is critical for the formation of osteoclasts by murine splenocytes incubated with M-CSF and RANKL in vitro. DPCPX inhibited osteoclast formation in a dose-dependent fashion (P = 0.0033); DPCPX inhibits osteoclastogenesis by directly acting on osteoclast precursors on day 0 (95.00 ± 2.309 vs. 24.00 ± 4.509 N = 3, P = 0.0002) compared to day 3 and 6 (95.00 ± 2.309 vs. 63.33 ± 14.53 N = 3, P = 0.0977) since removal of DPCPX by adding N6CPA after 3 and 6 days after the start of incubation did not reverse the inhibition of osteoclastogenesis (24.00 ± 4.509 N = 3 vs. 29.33 ± 2.333 N = 3, P = 0.3528). In contrast to their wild type controls splenocytes from A1 receptor KO mice formed almost no osteoclasts in response to MCSF and RANKL (120.4 ± 23.93 N = 5 vs. 21.40 ± 2.839 N = 5, P = 0.0034). The osteoclast number and function in the A1 KO mice did not appear to be normal as, in contrast to WT controls, osteoclasts in the A1 KO mice are not attached to bone and there is little bone resorption associated with these osteoclasts.


These results indicate that endogenously released adenosine, acting at adenosine A1 receptors, plays a critical role in the formation of osteoclasts and bone remodeling. Study presented above will probably lead to new therapeutic approaches in several diseases that are characterized by excessive bone resorption. These results suggest that adenosine A1 receptor antagonists may be useful in the treatment of such conditions as osteoporosis, prosthetic joint loosening and other conditions in which osteoclasts play a pathogenic role (eg Paget's disease).

Adenosine A2A Antagonism is Protective in a Model of Focal Cerebral Ischemia in the Rat

Pedata F., Gianfriddo M., Vannucchi M.G.1, Cipriani S., Giovannini M.G. and Melani A.

Department of Preclinical and Clinical Pharmacology, University of Florence, 50139 Florence, Italy; 1Department of Histology, University of Florence, 50139 Florence, Italy. felicita.pedata@unifi.it

Previous results indicate that the adenosine A2A antagonist SCH 58261 administered acutely immediately after focal ischemia reduces glutamate outflow and acute motor disturbance in the first hours after ischemia1.

The effect of a subchronic treatment of the A2A receptor antagonist, SCH58261, was studied in the same model of focal cerebral ischemia. Focal ischemia was induced by middle cerebral artery occlusion (MCAo)2. SCH58261 (0.01 mg/kg, i.p.) was administered 5 min, 6 hours and 15 hours after MCAo. Soon after ischemia, contralateral turning behavior was evaluated as the number of rotations per hour between 3 and 4 h after MCAo. In SCH58261- treated rats (n = 14), the number of rotations per hour was significantly reduced with respect to vehicletreated rats (n = 13) (mean ± S.E.: 116.9 ± 34.6 vs 795.4 ± 170.6, p < 0.0001). Twenty-four hour after MCAo, neurological deficit and ischemic brain damage were evaluated. SCH58261-treated rats (n = 14) showed significant improvement of the neurological score (mean ± S.E: 10.8 ± 0.4 vs 8.8 ± 0.5, p < 0.001) and reduction in the extent of the ischemic damage by 26% in the cortex (41.1 ± 2.8 mm3 vs 55.6 ± 3.9 mm3, p < 0.02) and by 45% in the striatum (12.8 ± 1.9 mm3 vs 23.2 ± 2.7 mm3, p < 0.01) with respect to vehicle-treated rats. The rats were then sacrificed for evaluation of phospho-p38 MAPK levels in the ischemic hemisphere. The phospho-p38 MAPK levels in the ischemic striatum of vehicle-treated rats (n = 5) were increased by 500% compared to the contralateral non ischemic striatum. In SCH58261-treated rats (n = 6), the phospho-p38 MAPK levels were significantly reduced by 70% in the ischemic striatum (p < 0.01) with respect to vehicle-treated rats. In the striatum and cortex the phospho-p38 MAPK immunopositive cells showed swollen and hypertrophyc cell bodies, as well as processes, and exhibited morphological features of activated microglia. Twenty-four hour after MCAo, astrocytes were found only in the corpus callosum. Activated microglia immunostained by OX-42 or isolectin B4 was present in the same cortical and striatal areas where phospho-p38 MAPK immunopositive cells were detected. SCH 58261 reduced phospho-p38 MAPK immunoreactivity in the striatum and cortex without changing the microglial cell morphology.

Results indicate that the protective effect of the adenosine antagonist SCH 58261 24 hours after ischemia is not due to reduced microglial activation but may involve inhibition of phospho-p38 MAPK and indicate that treatment with the A2A antagonist from the first to several hours after ischemia may be a useful therapeutic approach in cerebral ischemia. (Grant from University of Florence and Ente Cassa di Risparmio Firenze, Italy).

Adenosine A2A receptor (A2AR) antagonism prevents dermal fibrosis

Patricia Fernandez1, Sean Trzaska1, Carmen Montesinos2, Michael H Pillinger1, Allison B Reiss3, Bruce N Cronstein1 & Edwin SL Chan1

1New York University School of Medicine, New York, NY; 2University of Valencia, Valencia, Spain; 3Winthrop University Hospital, Mineola, NY Patricia.Fernandez@med.nyu.edu


Adenosine is a potent endogenous regulator of inflammation and tissue repair. Since the adenosine A2A receptor (A2AR) promotes dermal wound closure and increases dermal matrix deposition, we determined whether adenosine may also play a role in fibrosis in pathological conditions such as scleroderma. We therefore investigated the effect of A2AR activation on fibroblast collagen production and examined the role of A2AR in bleomycininduced dermal fibrosis, a model of scleroderma.


Primary human dermal fibroblasts (DFs) were incubated with the A2AR agonist CGS-21680 (1 nM–10 µM, 24 hrs) in the presence or absence of A2AR antagonist ZM241385 (1 µM). Collagen content in whole cell lysates and supernates was analyzed by Western blot and Sircol assay respectively. Collagen I and collagen III mRNA (10 hrs) were quantitated by real-time RT-PCR. Dermal fibrosis was induced with bleomycin (0.1 U sc qod × 18 days) in A2AR-deficient (A2AKO) versus wild-type (WT) littermate mice and in C57BL/6 mice treated with or without ZM241385 (50 mg/kg/day ip) and compared to PBS-treated controls. Dermal morphometric measurements and hydroxyproline content (reflecting collagen) were analyzed at sacrifice.


CGS-21680 treatment increased type I collagen in dermal fibroblast lysates by up to 151 ± 21% in a dose-dependent manner and this increase was reduced by coincubation with ZM241385 (1 µM, p < 0.04). Total collagen content was increased in supernates of CGS-21680 stimulated DFs to 176 ± 14% of control (n = 3, p < 0.01), and this increase was completely abrogated by ZM241385 (10 µM, n = 3, p = NS vs. control). CGS-21680 increased collagen I mRNA (146 ± 15% control, n = 4, p < 0.05) as much as TGF-β (10 ng/ml), with a lesser increase in collagen III mRNA (124 ± 11% control, n = 4, p = NS). Following bleomycin treatment, WT mice had greater punch biopsy thickness, skin-fold thickness and higher dermal tensile strength (p < 0.05 for all, data not shown) than A2AKO mice. Furthermore, dermal hydroxyproline content was higher in WT than A2AKO mice (27.4 ± 2.3 vs. 19.5 ± 0.7 µg/mg tissue, WT vs. A2AKO, respectively, n = 5, p < 0.005). Compared to vehicle-treated mice, ZM241385-treated mice were also protected from bleomycin-induced increases in morphometric measures as well as in dermal hydroxyproline content (26.2 ± 1.0 vs. 20.9 ± 1.0 µg/mg tissue, control vs. ZM, n = 5, p < 0.005).


Adenosine, released in response to hypoxia or cell injury, promotes, collagen production by human dermal fibroblasts via A2AR occupancy. More importantly, deletion or blockade of adenosine A2AR protects mice from bleomycin-induced dermal fibrosis, a model of such fibrosing conditions as scleroderma, and suggests a novel therapeutic use for A2AR antagonists for the prevention of fibrosis in the skin.

Adenosine A2A receptor activation and IgG-E. coli-immune complexes synergistically up-regulate IL-10 production in mouse macrophages

Balázs Csóka, Zoltán H. Németh, György Haskó

Department of Surgery, UMDNJ-New Jersey Medical School, Newark, NJ 07103 csokaba@umdnj.edu

IL-10 is produced by monocytes/macrophages, and it has important anti-inflammatory and immunomodulatory effects. This function of IL-10 is important for protecting the host from harmful effects of prolonged inflammatory responses in the context of microbial infection. Adenosine, an endogenous purine nucleoside, is formed at sites of metabolic stress associated with inflammation. Adenosine mediates its effects through four adenosine receptors (A1, A2A, A2B, A3), all of which are expressed on macrophages. Stimulation of adenosine receptors results in increased IL-10 production (∼2-fold) in macrophages activated by endotoxin. However, the effect of adenosine on IL-10 production in IgG-E.coli immune complex-treated macrophages is not known. In this study we found that IgG-E. coli-immune complexes increased IL-10 production approximately two-fold in peritoneal macrophages. Adenosine dramatically (∼8-10-fold) enhanced IgG-E. coli immune complex-activated IL-10 production, indicating a synergistic effect. The A1 receptor agonist CCPA and the A3 adenosine agonist IB-MECA failed to mimic the stimulatory effect of adenosine on immune complex-induced IL-10 production. However, both the selective A2A receptor agonist CGS-21680 and the non-selective agonist NECA increased IL-10 production, and CGS-21680 was the most potent. To further confirm the role of A2A receptors in promoting immune complex-primed IL-10 production, the abilities of selective adenosine receptor antagonists were examined in preventing the stimulatory effect of adenosine. We found that only the selective A2A antagonist ZM241385 reversed the effect of adenosine on immune complex-induced IL-10 production. In addition, we found that adenosine had no effect on immune complex-primed IL-10 production in A2A KO mouse macrophages, which also emphasizes the importance of A2A receptor ligation in IL-10 up-regulation. To investigate the intracellular mechanisms responsible for the augmented production of IL-10 by adenosine, we determined IL-10 mRNA levels using real-time PCR. We found that adenosine increased IL-10 mRNA levels in IgG-E. coli-immune complex-treated cells. We also examined the effect of adenosine on immune complex-induced IL-10 promoter activity by transfecting RAW264.7 cells with a construct in which luciferase expression is driven by the IL-10 promoter. We found that adenosine enhanced immune complex-induced IL-10 promoter activity, and this effect was even more enhanced when RAW264.7 cells were co-transfected with an A2A receptor protein overexpressing construct. Collectively, these results suggest that adenosine up-regulates IgG-E. coli immune complex-induced IL-10 production. This effect is mediated by A2A adenosine receptor activation, and is associated with increased IL-10 mRNA accumulation and enhanced IL-10 promoter activation. Moreover the synergistic effect of adenosine was more robust on IgG-E. coli-activated IL-10 production than on LPS-induced IL-10 production.

Adenosine A2A receptor activation reduces lung injury in trauma/hemorrhagic shock

György Haskó, Balázs Csóka, DaZhong Xu, Qi Lu, Zoltán H. Németh, Edwin A. Deitch

Department of Surgery, UMDNJ-New Jersey Medical School, Newark, New Jersey, USA haskoge@umdnj.edu

Hemorrhagic shock and resuscitation trigger a global ischemia/reperfusion phenomenon, in which various inflammatory processes critically contribute to the ensuing tissue damage. Adenosine is an endogenous nucleoside that is released during shock. Activation of adenosine A2A receptors can broadly inactivate inflammatory cascades. The current study was designed to evaluate the effect of A2A receptor activation on organ injury and inflammation in the setting of global ischemia/reperfusion elicited by trauma/hemorrhagic shock and resuscitation. Adult male Sprague-Dawley rats were subjected to a laparotomy (trauma) and 90 minutes of hemorrhagic shock or trauma/sham shock. The selective A2A receptor agonist CGS-21680 (2-p-(2-carboxyethyl) phenethylamino-5′-N-ethyl-carboxamidoadenosine; 0.5 mg/kg) or its vehicle was injected 30 min before shock or immediately after resuscitation. At 3 hours following resuscitation, animals were killed and tissue harvested for analysis. Lung permeability and pulmonary myeloperoxidase levels were used to quantitate lung injury. Intestinal injury was determined by histologic analysis of terminal ileum. Red blood cell (RBC) deformability was measured by a laserassisted ektacytometer. Pretreatment with CGS-21680 protected the lung but not the gut against shock-induced injury and prevented the shock-induced decrease in RBC deformability. Post-treatment with CGS-21680 ameliorated shock-induced lung injury but failed to prevent gut injury and preserve RBC deformability. A2A receptor agonists may represent a novel therapeutic approach in preventing organ injury following trauma/hemorrhagic shock.

Adenosine A2A Receptor Agonist Protects Mice from Compression-Induced Spinal Cord Injury

Yuesheng Li and Joel Linden

Cardiovascular Research Center, University of Virginia Health System, Charlottesville, Virginia 22908 yjl4n@virginia.edu

The adenosine A2A receptor (A2AR) has been suggested as a predominant anti-inflammatory adenosine receptor1. In our previous studies, we have shown that activation of the A2AR inhibits the activity of various inflammatory cells in vitro2, 3, 4 and in vivo5,6. Although inflammation clearly exacerbates ischemia-reperfusion injury (IRI) in heart, liver and kidney, it is not clear if this is also the case in spinal cord. We have developed a new system for scoring hindlimb locomotor dysfunction in mice, the mBBB scoring system. This system has improved our ability to evaluate compression-induced locomotor dysfunction following spinal cord injury (SCI) in mice7. We show that the A2AR selective agonist, ATL313, transiently applied during or after spinal cord IRI persistently reduces mouse SCI based on the new mBBB locomotor score and a previously validated scoring system, tBBB8. Female mice were subjected to compression-induced SCI after laminectomy and periodically evaluated over 6 weeks. A 1 × 2 mm region of spinal cord was compressed for 5 min with a 15 g weight applied to the dorsal surface of the cord at T12. ATL313 was administered IP for 5 minutes either before or just after injury, on the evening after surgery, and twice daily for the next 3 days. Compared to vehicle controls ATL313-treatment produced significant locomotor improvement that reached a plateau within 7 days and was sustained until the end of the experiment at 42 days. ATL313 did not influence SCI in A2AR knockout mice. Ongoing experiments with bone marrow chimera mice created by irradiation and bone marrow transplantation indicate that spinal cord protection by ATL313 is due to receptors on bone marrow-derived cells and not due to receptors on neurons or the vasculature. The results suggest that inflammation does exacerbate SCI following spinal cord IRI, and that A2A agonists reduce injury by reducing inflammation. A2A agonist therapy may be useful for treating patients subjected to spinal cord ischemia during the repair vascular tears or aneurisms.

Adenosine A2A receptor inactivation increases survival in polymicrobial sepsis

György Haskó1, Balázs Csóka1, Jeanette Wilmanski1, DaZhong Xu1, Qi Lu1, Catherine Ledent2, Edwin A. Deitch1, Pál Pacher3, Zoltán Spolarics1, and Zoltán H. Németh1

1Department of Surgery, UMDNJ-New Jersey Medical School, Newark, NJ 07103, USA; 2Institut de Recherche Interdisciplinaire en Biologie Humaine et Nucléaire, Université Libre de Bruxelles, B1070, Brussels, Belgium; 3National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD 20892-8115 haskoge@umdnj.edu

The mechanisms governing the impairment of bacterial clearance and immune function in sepsis are not known. Adenosine levels are elevated during tissue hypoxia and damage associated with sepsis. Adenosine has strong immunosuppressive effects, many of which are mediated by A2A receptors expressed on immune cells. We examined whether A2A receptors are involved in the regulation of immune function in cecal ligation and puncture-induced murine polymicrobial sepsis by genetically or pharmacologically inactivating A2A receptors. A2A receptor KO mice were protected from the lethal effect of sepsis and had improved bacterial clearance compared to WT animals. cDNA microarray analysis and flow cytometry revealed increased MHC II expression in A2A-inactivated mice, suggesting improved antigen presentation as a mechanism of protection. Apoptosis was attenuated in the spleen of A2A KO mice indicating preserved lymphocyte function. Levels of the immunosuppressive cytokines IL-10 and IL-6 were markedly lower following A2A receptor blockade. Similar to observations with A2A receptor KO mice, an A2A receptor antagonist increased survival even when administered in a delayed fashion. These studies demonstrate that A2A receptor blockade may be useful in the treatment of infection and sepsis.

Adenosine A2A Receeptor Levels in Different Peripheral Cells of Alzheimer's Disease Patients

Lorenza Galimberti1, Beatrice Arosio1, Carmen Calabresi1, Silvia Scurati1, Giorgio Annoni2 and Carlo Vergani1

1Department of Internal Medicine, Chair of Gerontology and Geriatrics, IRCCS Foundation, Ospedale Maggiore Policlinico, Mangiagalli e Regina Elena, University of Milan, Milan, Italy; 2Department of Clinical Medicine, Prevention and Medical Biotechnology, University of Milano-Bicocca, Milan, Italy lorenza.galimberti@unimi.it

Adenosine, released by cells and tissues, is an important endogenous purine neuromodulator which potently regulates inflammation. Its physiological effects are transduced through four different receptor subtypes (A1, A2A, A2B and A3) that are variably expressed on immune and inflammatory cells. All adenosine receptors are G-protein coupled receptors (GPCR) of the type 1 variety which act by modulating adenylyl cyclase(1). In particular the A2A receptor subtype seems to be implicated in neuroprotective and anti-inflammatory mechanisms(2); indeed the accumulation of extracellular adenosine in inflammatory and damaged tissues and the immunosuppressive properties of cAMP-elevating receptors indicate that signaling by A2A could be a possible natural mechanism of inhibition and/or termination of inflammation(3,4). The aim of this study is to investigate if an altered expression and/or function of A2A receptors, could contribute to the inflammatory/neurodegenerative mechanisms involved in Alzheimer's disease (AD). We analysed the expression of this receptor in platelets and Peripheral Blood Mononuclear Cells (PBMCs) from AD patients and sex- and age-matched healthy controls (HC); furthermore we analysed a T/C single nucleotide polymorphism (SNP) at position 1083 of exon 2 of the A2A gene that seems to be correlated with other neurodegenerative disorders. We found no statistical significative differences in the distribution of this SNP between AD and HC. By contrast the A2A mRNA levels in platelets were 25% higher in HC than AD (1.09 and 0.82 respectively; p = 0.000). To confirm these data we are analysing in a wider group of AD and HC the mRNA and protein expression of A2A in PBMCs. The preliminary results match with those obtained from platelets, suggesting a lower ability of AD subjects to terminate the inflammatory response and indicating a possible role of the A2A receptor in the pathogenesis of AD.

Adenosine A2A receptors modulate psychomotor activity and brain injury by distinct cellular mechanisms

Liqun Yu1, Qing-yuan Huang1, Nelson Rebola4, Hai-Ying Shen1, Eric Kirsten Rapp1, Yuan-Ji Day3, Jarrod Ferrara1, Joana E. Coelho1, Paula M. Canas4, Zhi-Hong Huang2, Darcie Taylor1, Michael Moskowitz2, Michael Schwarzschild2, Joel Linden3, Rodrigo A. Cunha4 and Jiang-Fan Chen1#

1Department of Neurology, Boston University School of Medicine, Boston, MA; USA 2Neuroscience Center and Department of Neurology, Mass. General Hospital, Boston, MA; USA 3Department of Internal Medicine, University of Virginia, Charlottesville, VA; USA 4Center for Neuroscience, Institute of Biochemistry, University of Coimbra, Portugal

The adenosine A2A receptor (A2AR) has recently emerged as a leading non-dopaminergic therapeutic target for Parkinson's disease for its ability to regulate motor activity. Furthermore, A2ARs influence brain injury outcome in variety of neurological disease models, presumably through modulation of glutamate release. Using forebrain neuronal-specific A2AR knockout (KO) mice, we here provide the first direct evidence that A2AR-mediated control of motor function and neuroprotection involve distinct cellular mechanisms. By crossing the floxed A2AR mice with the CaMKII-Cre transgenic line, we selectively depleted A2AR mRNA and protein in forebrain neurons to the background level of the global A2AR KO mice, as demonstrated by in situ hybridization, immunochemistry and receptor binding assays. This genetic deletion of A2ARs in forebrain neurons abolished the psychomotor effect of the A2AR selective agonist CGS21680 and antagonist KW-6002 and of the non-selective antagonist caffeine, and largely attenuated the pyschostimulant effect of cocaine. This demonstrates the key role of forebrain neuronal A2ARs in the modulation of psychomotor activity. In contrast, genetic deletion of the A2AR in forebrain neurons did not confer protection against ischemic brain injury by middle cerebral arterial occlusion or against MPTP-induced dopaminergic neurotoxicity, despite abolishing CGS21680-mediated presynaptic facilitation of glutamate release in forebrain A2AR KO mice. Furthermore, intracerebral ventricular administration of KW-6002 into forebrain A2AR KO mice reinstated neuroprotection against MPTP neurotoxicity. These results provide the clearest data yet that A2AR activity in forebrain neurons is critical to control psychomotor activity, but not for neuroprotection against brain injury, indicating that A2ARs modulate motor activity and brain damage by distinct cellular mechanisms. This opens up the new possibility of selectively manipulating A2AR's motor and neuroprotective effects by targeting different cellular elements.

Adenosine A2A Receptors Stimulate Collagen Production in LX-2 Cells via PKA, SRC and MAPK Cascade Signaling Pathway

Jiantu Che and Bruce N. Cronstein

Divisions of Clinical Pharmacology and Rheumatology, Department of Medicine, New York University School of Medicine, New York, NY 10016


We have previously reported that adenosine A2A receptor deletion protects mice from toxin-induced hepatic fibrosis and that adenosine A2A receptor occupancy promotes collagen production by hepatic stellate cells. We therefore determined the signaling pathways involved in adenosine receptor-mediated stimulation of a hepatic stellate cell line, LX-2 cells.


Soluble collagen was detected by Sircol assay according to the manufacturers directions. Type I and type III collagen mRNA expression was measured by real-time PCR. In addition, collagen I and III levels were detected semi-quantitatively by densitometric measurement of Western Blots. Phospho-Erk1/2 was detected by Western Blot.


Soluble collagen was stimulated both in supernatant and the cells by CGS21680, an adenosine A2A receptor agonist, at concentrations ranging from 0.01–10 µM (P < 0.005 or P < 0.001). Starting at 6 hours after treatment and reaching a peak at 24 hours. CGS21680 stimulated an increase in both mRNA and protein of collagens I and III (P < 0.001). ZM241385, an adenosine A2A receptor antagonist, completely blocked CGS21680-stimulated increases in soluble collagen (by 100% at 24 hours after treatment, P < 0.001) as well as collagen I (by 100% at 24 hours after treatment, P < 0.001) and III (by 100% at 24 hours after treatment, P < 0.001) with a relative potency profile consistent with blockade of an A2A receptor (maximal effect at 1 µM). Furthermore, the effect of CGS21680 on stimulation of soluble collagen was attenuated by the addition of Peptide Inhibitor of protein kinase A (PKAI, by 30%, 1 µM, P < 0.005), Src inhibitor II (SRCI, by 28% P < 0.005), MEK-MAPK inhibitor U0126 (by 24%, P < 0.005), the ERK MAPK inhibitor PD98059 (by 37%, P < 0.001), and the p38 MAPK inhibitor SB202190 (by 35%, P < 0.001). Real-time PCR and Western blot analysis demonstrated that PKAI, SRCI, U0126 and PD98059 inhibited CGS21680-stimulated increases in both mRNA and protein of collagen I (by 50%, 48%, 44% and 42% respectively, P < 0.001), whereas SB202190 did not. In contrast, SB202190 inhibited CGS21680-stimulated type III collagen production (by 32%, P < 0.005), which was independent of PKA, Src and ERK MAPK signaling pathway. The effect of PKAI, SRCI, U0123 and PD98059 on CGS21680-stimulated collagen production in LX-2 cells was associated with significantly decreased ERK1/2 phosphorylation.


These results further indicate that adenosine A2A receptors increase collagen production in LX-2 cells via a PKA-Src-MEK-ERK MAPK cascade or p38 MAPK signaling pathway. Moreover, these results provide further evidence that adenosine A2A receptor antagonists may be used to prevent hepatic fibrosis. Finally, prior studies indicate that coffee drinking prevents hepatic cirrhosis and our studies suggest that the mechanism by which coffee drinking prevents hepatic cirrhosis is via blockade of adenosine A2A receptors in the liver.

Adenosine A2b receptor agonist mimics postconditioning: characterization of a selective adenosine A2b receptor agonist in a rabbit infarct model.

Thomas Krahn1, Barbara Albrecht1, Ulrich Rosentreter2, Michael V. Cohen3, Natalia Solenkova3, James M. Downey3

From Depts. of Pharmacology1 and MedicinalChemistry2, Pharma Research Center, Bayer Health Care, Wuppertal, Germany; Dept. of Physiology3, University of South Alabama, Mobile, AL USA Th.Krahn:Thomas.Krahn@bayerhealthcare.com

There is a continuing clinical need for an agent given after the onset of ischemia that will protect the heart. Herein we present evidence that a specific agonist for the adenosine A2b receptor is cardioprotective when administered just prior to termination of 30 min of regional ischemia in rabbit hearts. The role of A2b receptors has been largely unexplored in adenosine-mediated cardioprotection because of the lack of specific A2b receptor agonists. Bayer Healthcare was the first to synthesize specific non-adenosine like A2b receptor agonists.1 These A2b specific agonists were characterized in CHO cells expressing human A1, A2a or A2b receptors. Here we report on BAY A2b, a specific A2b agonist; EC50 values for receptor activation were >10.000 nM for A1, >10.000 nM for A2a and 3 nM for A2b receptors. Major aspects concerning synthesis and selectivity of non-adenosine like A2b receptor agonists were also reproduced by other groups.2

BAY A2b was given to open-chest rabbits experiencing 30 min of regional ischemia and 3 h of reperfusion. Rabbits were anesthetized with sodium pentobarbital (30 mg/kg) and mechanically ventilated with 100% oxygen. The heart was exposed through a left thoracotomy and a ligature was passed under a coronary branch to create ischemia. Drug was given intravenously 10 ug/kg over1min starting5min prior to reperfusion and again 15 min after reperfusion. The heart was removed after 3 h of reperfusion. The risk zone was stained with fluorescent microspheres and the heart was cut into 2 mm slices and infarct size was determined by tetrazolium staining.3

No adverse hemodynamic effects were seen with the agent. A better than 50% reduction of infarct size was seen and was comparable to that in a third group receiving postconditioning4, an established cardioprotective intervention where the occluded artery is intermittently opened and closed for four 30-second cycles at the end of the ischemic insult. The A2b agonist BAY A2b is equivalent to postconditioning in its potency. In conclusion we have synthesized specific A2b receptor agonists and have shown that an A2b receptor agonist can reduce infarct size by application after the onset of ischemia. Open image in new window

Adenosine and heme oxygenase 1 anti-inflammatory mechanisms are intimately interwoven in a regulatory loop involving the adenosine A2a receptor and carbon monoxide

A. Haschemi1,2, L.E. Otterbein2, R. Marculescu1, S.C. Robson3, F.H. Bach2 and O. Wagner1

1Department of Laboratory Medicine, Medical University of Vienna, Austria. 2Department of Surgery and 3Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, USA. arvand@gmx.net


Adenosine and heme oxygenase 1 (HO-1) are crucial regulatory molecules that participate in inflammation and immunity. HO-1 is a protective, inducible enzyme responsible for intracellular heme degradation that exerts a wide range of anti-oxidative, anti-inflammatory and immunomodulatory actions. These are chiefly mediated by the respective heme degradation products viz. carbon monoxide (CO), biliverdin/bilirubin and iron with apo-ferritin induction. In a similar manner, CD39 and other ecto-nucleotidases catalyze the phosphohydrolysis of extracellular nucleotides to generate the respective nucleosides. Adenosine is an endogenous nucleoside that is usually present at low concentrations in the extracellular space. However, with stress or in several patho-physiologic conditions, e.g. acute inflammatory responses, there may be dramatic increases in extracellular levels of all purinergic mediators. Adenosine modulates the kinetics and magnitude of immune responses via four different G protein-coupled receptors (A1, A2a, A2b and A3) that are differentially expressed on a variety of cells. In spite of their apparent lack of connection to one other, there are striking similarities in the anti-inflammatory profiles of adenosine and HO-1 and its products (as detailed above).


To elucidate possible cross talk between these two highly conserved molecules and/or catalytic products in the regulation of inflammatory and immune responses.

Results and Discussion

Adenosine exposure induces high level HO-1 expression in the macrophage cell line RAW264.7. This effect, appears to be mediated via the A2A receptor. In a reciprocal manner, high HO-1 expression remarkably sensitizes the macrophages to adenosine and potentiates the anti-inflammatory effects of adenosine and its analog 5′-N-ethylcarboxamidoadenosine (NECA) in an in vitro model of LPS activation. Adenosine mediated inhibition of LPS-induced TNF-α expression, is in part, HO-1 dependent as blockade of HO-1 via siRNA abrogates the effects of adenosine on TNF-α. We demonstrate that HO-1 boosts the expression of the anti-inflammatory A2a receptor and that CO can mimic and substitute for HO-1 in this response in RAW264.7 cells and mouse bone marrow derived macrophages which was not observed with biliverdin, one of its other products. Moreover, CO exposure prevents NECA induced down regulation of A2A receptors supporting the concept that HO-1 and adenosine are interrelated.


We demonstrate intimate cross talk between adenosine and HO-1/products. Based on these observations we can envision feedback loops, which may play a key role in the orchestration and resolution of inflammation.

Adenosine deaminase deficiency promotes dermal fibrosis

Patricia Fernandez1, Sean Trzaska1, Janci L Chunn2, Michael R Blackburn2, Bruce N Cronstein1 & Edwin SL Chan1

1Department of Medicine, New York University School of Medicine, New York, NY; 2Department of Biochemistry and Molecular Biology, University of Texas-Houston Medical School, Houston, TX Patricia.Fernandez@med.nyu.edu


Adenosine deaminase (ADA) catalyzes the deamination of adenosine to inosine. Recent studies indicate that chronic adenosine elevations lead to pulmonary fibrosis in ADA-deficient mice, suggesting a profibrotic role for adenosine in the lung [1]. We have also found that in a model of bleomycin-induced dermal fibrosis that mice lacking adenosine A2A receptors were protected from developing skin fibrosis. We therefore tested the hypothesis that chronic elevations of adenosine concentrations, due to either injury (as in the bleomycin-induced model of dermal fibrosis) or ADA deficiency, leads to dermal fibrosis.


Male ADA deficient mice (ADA KO) were supplemented with PEG-ADA enzyme therapy for the first three weeks after birth. After the last enzyme injection, animals were mantained without PEG-ADA for two weeks and then sacrificed. Dermal morphometric measurements were assessed on freshly excised skin. In some experiments mice were treated with the adenosine A2A receptor antagonist ZM241385 (50 mg/kg/day IP) for the last 8 days prior to sacrifice when the mice were no longer treated with PEG-ADA.


ADA KO mice showed significant increases in dermal thickness (128.7 ± 2.0%, n = 5, p < 0.001), skin-fold thickness (140.6 ± 2.8%, n = 5, p < 0.001) and breaking tension (136.8 ± 5.3%, n = 5, p < 0.001) compared with their wild type (WT) littermates. Dermal hydroxyproline content, a measure of collagen content, was also increased by 157.8 ± 2.8% (n = 5, p < 0.001) in ADA KO compared to the WT mice. To determine whether the adenosine A2A receptor was involved in the dermal fibrosis observed in the ADA KO mice, we treated ADA KO mice with the A2A receptor antagonist ZM241385. After ZM241385 treatment, the skin of treated mice was not as thick (27.2 ± 8.3% decrease, n = 4, p < 0.05), skin-fold thickness was 21.7 ± 7.4% less (n = 4, p < 0.05), and dermal tensile strength was reduced by 48.9 ± 14.1% (n = 4, p < 0.05) compared with ADA KO mice treated with vehicle. Furthermore, dermal hydroxyproline content was decreased (61.4 ± 13.5%, n = 4, p < 0.001) in ZM treated ADA KO mice compared with vehicle-treated ADA KO mice.


The marked increase in adenosine levels seen in other tissues in adenosine deaminase deficiency may act similarly on skin to promote fibrogenesis. Pharmacological protection of ADA deficient mice with A2A receptor antagonist administration suggests a role for A2A receptors in the modulation of dermal fibrogenesis.

Adenosine inhibits the release of arachidonic acid and its metabolites (AAM) in activated human peripheral mononuclear cells

Sándor Sipka1, Ildikó Kovács1, Sándor Szántó1, Gyula Szegedi2, László Brugós3, Gé za Bruckner5, A. József Szentmiklósi4

13rd Department of Internal Medicine, University of Debrecen, Hungary 2Research Group of Autoimmune Diseases, Hungarian Accademy of Sciences, Debrecen, Hungary 3Department of Pulmonology, University of Debrecen, Hungary 4Department of Pharmacology and Pharmacotherapy, University of Debrecen, Hungary 5Division of Clinical Nutrition, University of Kentucky, USA sipka@iiibel.dote.hu

The effects of adenosine (Ado) and subtype-specific activators of adenosine receptors (A1, A2A, A2B and A3) were studied on the release of arachidonic acid and its metabolites (AAM) from human peripheral mononuclear cells (monocytes). In the cells activated by protein kinase C- specific phorbol ester (phorbol 12-myristate 13-acetate) and Ca2+-ionophore (A-23187), adenosine and two subtype specific receptor agonists, CPA (A1) and CGS-21680 (A2A) induced concentration dependent inhibitions in the release of AAM, whereas the stimulation of A2B and A3 receptors were ineffective. The rank order of potency in the inhibition of AAM release was as follows: CGS-21680 = CPA > adenosine > NECA (in the presence of ZM-24185 and DPCPX as A2A and A1 adenosine receptor antagonists) = IB-MECA. Adenosine inhibited the release of AAM only at and over the concentration of 10−4 M, whereas the inhibitory effect of A1 and A2A receptor specific agonists appeared at the range of 10−7 M. It can be concluded that adenosine physiologically may not have a significant effect on the AAM release of circulating monocytes, but in pathological conditions, where the local Ado concentrations increases, this nucleoside by the activation of A2A and A1 receptors can exert its antiinflammatory action via the decrease in the proinflammatory AAM production.

Adenosine is induced during peritonitis and downregulates cytokine production and leukocyte recruitment

Sigal Nakav1, Nadav Y. Ziv1, Boris Rogachev2, Julia Mazar1, Cidio Chaimovitz2, Moshe Zlotnik2 and Amos Douvdevani1,2.

1Department of Clinical Biochemistry and 2Department of Nephrology, Soroka Medical Center, Ben-Gurion University of the Negev, Beer-Sheva, Israel. sigs@bgu.ac.il

Adenosine is an endogenous immunomodulator that has been shown to exhibit anti-inflammatory and immunosuppressive effects. These anti-inflammatory effects depend mainly on ligation with its cell-surface receptors subtypes: A2A receptor (A2AR), A2B receptor (A2BR), A1 receptor (A1R) and A3 receptor (A3R) all of which are G-protein coupled. The generation of extracellular adenosine involves phosphohydrolysis of adenine nucleotide intermediates and is regulated by two enzymes, nucleotidase triphosphate dephosphorylase (CD39) which converts ATP to AMP, and 5′ ectonucleotidase (CD73) which converts AMP to adenosine. Peritoneal mesothelial cells (PMC) form a monolayer that covers the peritoneal membrane. Their location between the peritoneal cavity and peritoneal blood vessels gives them a key role in intraperitoneal immune defense. Following stimulation with inflammatory cytokines and bacterial products, mesothelial cells express adhesion molecules and produce various cytokines and other pro-inflammatory mediators.

The aim of the present study was to elucidate the regulatory role of adenosine during peritonitis and to assess the regulation of CD39, CD73 on PMC during the inflammatory processes.

In a mouse model of E. coli-induced peritonitis we found a gradual increase of adenosine levels which peaked at 24 hours then gradually declined up to 72 hours from inoculation. The intra-peritoneal influx of leukocytes after inoculation was blocked by the A2AR agonist CGS-21680. In inoculated mice, the A2AR agonist also caused a significant decrease in sera and peritoneal levels of TNF-α and IL-6 as compared to untreated mice. Analysis of PMC mRNA and protein levels showed that both CD39 and CD73 levels increased ∼3 fold higher than normal at the initial phase of inflammation and decreased at the resolution phase.

These data suggest that adenosine is a potent regulator of peritoneal inflammation and the upregulation of both CD73 and CD39 in the initial phase of peritonitis is responsible for the increase of peritoneal adenosine levels.

Adenosine modulates the release of catecholamines from rat carotid body chemoreceptor cells through an interaction between D2 dopamine receptors and A2B adenosine receptors

Conde SV, Obeso A and Gonzalez C

Department of Biochemistry, Molecular Biology and Physiology, IBGM, Faculty of Medicine, University of Valladolid, CSIC, Valladolid, Spain. svconde.farm@fcm.unl.pt

Carotid bodies (CB) are major peripheral chemoreceptor organs sensing changes in blood O2 responding by generating action potentials at the carotid sinus nerve (CSN), which are integrated in the brainstem to induce a hyperventilatory compensatory response. Hypoxia, the physiological CB stimulus increases the release of dopamine 7 and adenosine 1,2 from rat CB. Adenosine is an excitatory neurotransmitter at the CB, increasing CSN electrical activity and promoting hyperventilation through the activation of A2 adenosine receptors 4,5,6. Recently, it has been described that caffeine inhibits the basal and evoked release of catecholamines (CA) from rat carotid body chemoreceptor cells (CBCC) through an action on A2B adenosine receptors 3. The present work was performed in order to investigate possible interactions between D2 dopamine and A2B adenosine receptors responsible for the modulation of CA release in rat CBCC. Experiments were performed in CB removed from 3 months old Wistar rats. The effect of adenosine A2B and dopamine D2 receptor agonists and antagonists applied alone or conjunctly were studied on the basal and evoked release (10% O2) of CA from CBCC. Dose response curves for these A2B and D2 receptor agonists and antagonists were performed. NECA, an A2 receptor agonist (0.1–100 µM), increase the basal and 10% O2-evoked release of CA from rat CB in a dose-dependent manner. Haloperidol (0.01–10 µM) and sulpiride (0.1-10 µM), D2 dopamine antagonists, increase in a dose-dependent manner the basal and/or evoked release CA from CBCC. Propylnorapomorphine, D2 dopamine agonist (0.2–200 nM), induced a dose-dependent decrease in the basal and evoked release of CA from rat CBCC. NECA (10 µM) when applied in association with haloperidol (0.01–10 µM) potentiates the effect of this D2 antagonist, moving its doseresponse curve to the left. Sulpiride (1 µM), reversed the inhibitory effect of caffeine on basal and stimulus induced release of CA from rat CBCC. Therefore, our results suggest that an interaction between A2B adenosine and D2 dopamine receptors could exist in CBCC, contrarily to the described in the CNS between A2A and D2 receptors, modulating the release of CA.

Supported by MEC (Spain) Project BFU2004-06394/BFI, Red Respira (Iciii/RTIC C03/011) and by JCyL grant VA 106A05. SV Conde is funded by a PhD grant from FCT (Portugal).

Adenosine produced via the CD73/ecto-5-nucleotidase pathway has no impact on erythropoietin production but is associated with reduced kidney weight

Burcin Özüyaman1, Ulrich K.M. Decking1, Zhaoping Ding1, Anja Buchheiser1, Patrjcya Koszalka1, Axel Gödecke1, Norbert Braun2, Herbert Zimmermann2, Jürgen Schrader1

1 Institut für Herz- und Kreislaufphysiologie, Heinrich-Heine-Universität Düsseldorf, Germany 2 Biozentrum der J.W. Goethe-Universität, Institut für Zellbiologie und Neurowissenschaft, Frankfurt am Main, Germany decking@uni-duesseldorf.de

CD73/ecto-5-nucleotidase which catalyzes the conversion of AMP to adenosine has been implicated in vascular homeostasis. In kidneys, exogenous adenosine is known to dose-dependently act as a vasodilator and constrictor and has been shown to enhance erythropoietin production. We therefore hypothesized that CD73-derived adenosine promotes erythropoietin production and determines basal kidney perfusion.

CD73 knockout mutants recently generated in our laboratory were compared to wild type controls. CD73 local expression was assessed by enzyme histochemistry and immunocytochemistry, CD73 activity determined by a phosphate assay, adenosine determined by HPLC, erythropoietin expression by real-time quantitative PCR, erythropoietin plasma concentration by an immunoassay, and renal plasma flow and glomerular filtration rate by PAH and FITC-inulin, respectively.

Of all organs investigated, kidneys showed the most prominent CD73 activity that was preferentially located in peritubular fibroblasts of the renal cortex and the glomerular mesangium. In the absence of CD73, alkaline phosphatase remained unchanged but tissue adenosine was reduced both under control conditions (by 76%) and during normobaric hypoxia (by 72%). Despite the loss of CD73 activity and substantial reduction in adenosine, EPO mRNA and plasma protein concentrations did not differ between WT and CD73−/− under basal conditions and following normobaric hypoxia (8% O2) and carbon monoxide (0.1% CO) inhalation (both for 4 h). Although there was no difference in blood pressure and urine flow volume, the average weight of the two kidneys was reduced by 21% in the knockout (WT: 7.17 ± 1.18 mg/g body wt; CD73−/−: 5.70 ± 1.91 mg/g body wt). Measurement of renal plasma flow (RPF) and glomerular filtration (GFR) revealed no significant difference when related to the respective kidney weights.

Conclusion: The extracellular CD73 pathway is a major source of adenosine in the kidney. CD73 derived adenosine thas no impact on erythropoietin production under basal conditions and after hypoxic challenge but may influence kidney growth.

Adenosine receptor activation protects against the development of diabetes in mouse models of type 1 diabetes

Zoltán H. Németh1, Balázs Csóka1, David Bleich2, Csaba Szabó1, Edwin A. Deitch1, György Haskó1

1Department of Surgery, 2Department of Medicine, UMDNJ-New Jersey Medical School, Newark, NJ 07103, USA nemethzo@umdnj.edu

Type 1 diabetes or insulin-dependent diabetes mellitus (IDDM) is a disease affecting 1.2 million patients in the United States with 12,000 new cases diagnosed each year. The disease is characterized by the specific destruction of insulin-producing β-cells in the pancreatic islets of Langerhans by the immune system. The islets are invaded by immune cells, particularly by macrophages and T cells, and these cells are cytotoxic to islet β-cells, in part by generating cytokines and free radicals. Currently there is no available treatment to prevent or cure type 1 diabetes. Based on evidence that adenosine receptor occupancy suppresses inflammatory cytokine production, we hypothesized that selective stimulation of adenosine receptors using the potent and stable agonist 5′-N-ethylcarboxamidoadenosine (NECA) would attenuate the course of diabetes in multiple-low-dose streptozotocin (MLDS) treated and in non-obese-diabetic (NOD) mice. In the MLDS model, male CD-1 mice were treated with 40 mg/kg streptozotocin (STZ) for 5 consecutive days simultaneously with NECA or its vehicle, and after the cessation of STZ treatment NECA or vehicle treatment was continued for 21 additional days. We found that 0.03 mg/kg NECA treatment prevented the MLDS-induced hyperglycemia (blood sugar levels on day 21 in mg/dl: control: 410.3 ± 60, NECA: 222.8 ± 50.9, mean T SEM, p < 0.05). The decrease in blood sugar levels was associated with preservation of pancreatic insulin content. To assess the extent of inflammation in the pancreas, we measured the pancreatic levels of tumor necrosis factor-alpha (TNF-α), macrophage inflammatory protein 1-alpha (MIP-1α), interleukin-12 (IL-12), and interferon-gamma (IFN-γ). Pancreatic contents of all four cytokines were significantly lower in mice treated with NECA than in vehicle-treated mice as shown in the table below.

Cytokine Level (pg/mg prot.)


NECA (0.03 mg/kg)


1.45 ± 0.33

0.62 ± 0.03**


2.87 ± 0.91

0.47 ± 0.02**


20.19 ± 6.88

5.27 ± 0.05*


0.37 ± 0.13

0.09 ± 0.02*

Results are means ± SEM (n = 10). *p < 0.05, **p < 0.01 versus vehicle-treated mice.

We next examined the effect of NECA in NOD mice, a genetic model of type 1 diabetes that has more in common with human IDDM than the chemically induced MLDS model. NOD mice were given a single dose of cyclophosphamide (200 mg/kg) and then daily injections of NECA (0.01 mg/kg). Even this low dose of NECA significantly decreased the blood glucose level in NOD mice (vehicle: 318.5 ± 188 mg/dl, NECA: 155.8 ± 76 mg/dl, mean ± SEM, p < 0.05). Taken together adenosine receptor stimulation might have great therapeutic potential in the treatment of humans suffering from IDDM.

Adenosine receptors in colon carcinoma tissues and colon tumoral cell lines: focus on the A3 adenosine receptors

1Stefania Gessi, 1Stefania Merighi, 1Katia Varani, 1Elena Cattabriga, 1Annalisa Benini, 2Carlo Feo, 3Edward Leung, 3Stephen MacLennan, and 1Pier Andrea Borea.

1Department of Clinical and Experimental Medicine, Pharmacology Unit and Interdisciplinary Center for the Study of Inflammation, Italy; 2Department of Surgery, Anesthesiology and Radiology and 3King Pharmaceutical, Cary, North Carolina. gss@unife.it

The present study was undertaken to investigate the presence of adenosine receptors on human colon cancer and to evaluate the functional effect of these receptors on colon cancer cell biology. Therefore the expression of the A1, A2A, A2B and A3 subtypes by means of quantitative real time RT-PCR and binding studies in both resected colon cancer tissues from patients undergoing surgery and HT29, DLD1 and Caco2 colon carcinoma cell lines have been analysed. Our results show that in human colon cancer tissues the density of A1 and A2A subtypes was quite low in comparison with that of the A3 receptor. Analogous results were found in carcinoma cell lines. [3H]MRE 3008F20 reveals an affinity and binding capacity very similar in all the cell lines investigated. Moreover the effect of adenosine and of selective adenosine antagonists for A1, A2A, A2B and A3 subtypes on cell growth have been investigated. The results of this work suggest that adenosine modulated cell proliferation through the involvement of the A3 subtype.

Adenosine receptors in exocrine pancreas

Ivana Novak, Susanne E. Hede and Mette R. Hansen

August Krogh Building, Institute of Molecular Biology and Physiology, University of Copenhagen Copenhagen, Denmark. inovak@aki.ku.dk

Pancreatic acini secrete ATP in response to cholinergic stimulation. Recent experiments show that pancreatic juice also contains nucleotidases, CD39 and CD73. Thus apart from P2 receptors, also adenosine receptors could be potential regulators of exocrine secretion. The aim of the present study was to determine whether pancreatic ducts, which normally secrete HCO3 rich fluid, possess functional adenosine receptors.

Pancreatic ducts were obtained from collagense digests of rat pancreas. Electrical activity in single isolated ducts was monitored using whole-cell nystatin patch-clamp method (1). The membrane voltage (Vm) was continuously monitored during experiments in zero current-clamp mode. Periodically, voltage was clamped and the whole-cell current and the total conductance (Gt) were measured. In another set of experiments, the Fura-2 method was used to estimate [Ca2+]i, in pancreatic ducts. In addition, RT-PCR analysis was carried out on isolated ducts and whole pancreas. In about half of the ducts studied, adenosine (1–100 µM) had no effect on resting Vm of about −55 mV. However, in the other half of the ducts adenosine markedly and reversibly depolarized Vm by 10–40 mV, and also increased whole cell membrane conductances (Gm) by about 30%. In such ducts lowering of the extracellular Cl-concentrations led to a further depolarization of Vm, indicating activation of a Cl- conductance. In another set of experiments, the effect of adenosine on intracellular Ca2+ of isolated single ducts was monitored. However, adenosine had no significant effects in the Fura-2 ratio, indicating that intracellular signals were not mediated by Ca2+ signalling. RT-PCR analysis showed that the isolated ducts have transcripts for the following adenosine receptors: A1, A2a, A2b and A3.

In summary, the present study shows that some pancreatic ducts express functional adenosine receptors that lead to opening of Cl- channels and possibly to the production of HCO3 rich fluid. Since Ca2+ signals were not observed, it is likely that A2a or A2b receptors regulate CFTR Cl- channels in some pancreatic ducts. The projects were supported by the Danish Medical and Science Research Councils, the Augustinus Foundation and the Lundbeck Foundation.

Adenosine role in the formation of hepatic fibrosis

Zhongsheng Peng1, Simon Robson2, Linda Thompson3 and Bruce N. Cronstein1

1NYU School of Medicine, New York, NY; 2Beth Israel-Deaconess Medical Center, Boston, MA; 3Oklahoma Medical Research Foundation


Adenosine is a potent endogenous regulator of tissue repair released from injured cells and tissues. Hepatic fibrosis results from chronic and acute hepatic injury and we have previously reported that adenosine, acting at A2a receptors, plays a role in hepatic fibrosis. The enzymes involved in formation of adenosine following hepatic injury have not been characterized.


Mice were treated with PBS, CCl4, thioacetamide or ethanol, sacrificed and their livers harvested. Liver slices were incubated in medium for 24 hrs before adenosine concentration in the supernatant was measured by HPLC. Hepatic fibrosis was induced with thioacetamide in CD39 (nucleoside triphosphate phosphohydrolase) knockout mice, CD73 (ecto-5′nucleotidase) knockout mice, CD39/73 double knockout mice and C57BL/6 control mice and was quantified by digital analysis of picrosirius red stained slides and hydroxyproline content.


Thioacetamide, carbon tetrachloride and ethanol treatment led to a marked increase in adenosine in supernates of hepatic slices (419.4 ± 22.8 nM, 492.1 ± 68.2 nM, 694.9 ± 35.8 nM, respectively, vs 274.0 ± 14.7 nM in PBS-treated mice, n = 5 and P < 0.05 for all). Hepatic hydroxyproline content was significant decreased in CD39/73 knockout mice compared with C57BL/6 mice (0.31 ± 0.06 µg/ml versus 0.40 ± 0.08 µg/ml, P < 0.05). Hepatic fibrosis (%fibrotic area/hepatic slice area) was decreased in CD39KO (0.26 ± 0.12%, p < 0.05), CD73KO (0.31 ± 0.21%, p = NS) and CD39/CD73KO mice (0.29 ± 0.09%, p < 0.05) as compared to C57BL/6 (0.42 ± 0.17%).


These results are consistent with the hypothesis that following hepatic injury adenosine is produced extracellularly from adenine nucleotides and that the adenosine released promotes hepatic fibrosis. We conclude that inhibition of adenosine production or interaction with its receptor may help prevent hepatic fibrosis.

Adenosinergic Mechanisms Contribute to Individual Differences in the Effects of Sleep Deprivation on Psychomotor Performance and the EEG

Landolt HP1,2, Rétey JV1, Adam M1, Gottselig JM1, Khatami R1, Dürr R1, Achermann P1,2

1Institute of Pharmacology & Toxicology, University of Zürich, Zürich, Switzerland 2Center for Integrative Human Physiology, University of Zürich, Zürich, Switzerland landolt@pharma.unizh.ch

Sleep deprivation impairs performance on neurobehavioral tasks and modulates the electro-encephalogram (EEG) in wakefulness and sleep. The changes in performance and the EEG are assumed to reflect a sleep-wake dependent, homeostatic process of sleep regulation1. Our observation that the adenosine receptor antagonist, caffeine, attenuates the EEG correlates of sleep homeostasis in wakefulness and sleep is consistent with the hypothesis that the adenosinergic system is prominently involved in sleep-wake regulation2. Recent research highlights the importance of studying individual differences in wakefulness and sleep as a novel approach to gain insights into physiological mechanisms underlying human sleep-wake regulation3. We recently found that genetic variability in the adenosinergic system contributes to inter-individual differences in the waking and sleep EEG4. Here, we hypothesized that adenosinergic mechanisms also play a role for individual differences in neurobehavioral performance decline associated with prolonged wakefulness. We investigated psychomotor vigilance task (PVT) performance and antero-posterior EEG power gradients during wakefulness and sleep following one night of sleep deprivation. The PVT is a visual reaction time task that is sensitive to sleep-deprivation induced impairments in sustained vigilant attention. Subjectively caffeine sensitive (n = 12) and insensitive (n = 10) young men (age range: 20–30 years) underwent two 40-hour waking periods. After 11 and 23 hours waking they received caffeine (200 mg per dose) and placebo in double-blind, cross-over fashion. The PVT and the EEG were assessed in 14 sessions at 3-hour intervals. To quantify the effect of prolonged waking, the data from three test sessions that occurred at analogous time of the day (at 11:00, 14:00 and 17:00 h) before and after the night without sleep were averaged. The data were analyzed with repeated-measures analyses of variance (rANOVA) and Spearman's rank correlation analyses. In the placebo condition, optimal PVT performance (fastest 10th percentile of speed) was more impaired by sleep deprivation in caffeine sensitive men than in caffeine insensitive men (p < 0.04). This difference between the groups was not significant in the caffeine condition (p > 0.05). In the placebo condition, sleep deprivation enhanced a fronto-occipital EEG power gradient (fronto-central vs. parieto-occipital bipolar EEG derivations) during waking (6.5–8.5 Hz range) and non-rapid-eye-movement (nonREM) sleep (<1 Hz range) in caffeine sensitive subjects when compared to caffeine insensitive subjects (p < 0.01 and p < 0.001). After caffeine these differences in the regional EEG power distribution between the groups were not significant (p > 0.05). The sleep-deprivation and caffeine-induced changes in PVT performance, as well as in the EEG during wakefulness and nonREM sleep were significantly correlated on an individual basis. These results indicate that adenosinergic mecha-nisms contribute to individual differences in the vulnerability to sleep-deprivation induced changes in psychomotor function and in the EEG. Elucidation of these mechanisms will help in the identification of persons at risk for impaired performance following sleep deprivation and in the develop-ment of agents to reduce the detrimental effects of prolonged waking on cognition and behavior. Research supported by the Swiss National Science Foundation (Grant # 3100-067060.01).

Agonist-induced trafficking of the P2Y1 receptor quantified with the novel radioligand, [32P]MRS2500.

D. Houston, D.M. Bourdon, A.D. Qi, R.A. Nicholas, T. K. Harden.

Dept. Pharmacology, University of North Carolina School of Medicine, Chapel Hill, NC, USA. dhouston@med.unc.edu

G-protein coupled receptors are responsible for transducing signals from a variety of extracellular hormones, nucleotides and neurotransmitters to the intracellular machinery. Subsequent to agonist activation, GPCR signaling can be terminated by desensitization followed by internalization into intracellular compartments. The occurrence and mechanisms of agonist-induced internalization of the P2Y family of GPCRs are poorly understood, primarily due to a lack of tools available for studying natively expressed receptors. Recently, we synthesized a high affinity, high specific radioactivity radioligand, [32P]MRS2500, for the P2Y1 receptor (P2Y1-R) and established its utility for detecting and quantifying P2Y1-R in a variety of mammalian tissues. An intact cell binding assay was established using this ligand and applied to measure agonist-induced loss of [32P]MRS2500 binding sites in mammalian cell lines and intact platelets. Treatment of canine kidney (MDCK) cells with the P2Y1-R agonist 2MeSADP resulted in a 50% decrease in the number of surface [32P]MRS2500 binding sites. This decrease occurred with a ± 1/2 of approximately 5–10 minutes and reached a steady-state within 30 minutes that was maintained for up to one hour. The human P2Y1 receptor stably expressed in MDCK cells (P2Y1-MDCK) exhibited similar kinetics for agonist-induced internalization. The extent of agonist-promoted loss of surface P2Y1-R in P2Y1-MDCK cells was dependent on concentration of agonist, and internalization was inhibited almost completely by pretreatment of cells with 450 mM sucrose, an inhibitor of clathrin-mediated endocytosis. The initial loss in surface P2Y1-R binding sites in P2Y1-MDCK cells was completely reversed within 15 minutes after removal of agonist from the medium. In human platelets, agonist-induced internalization occurred with much faster kinetics, with the maximal extent of internalization occurring within five minutes. Taken together, these data indicate that the P2Y1-R undergoes clathrin-mediated endocytosis in response to agonist stimulation.

Agonists of the Adenosine A2a Receptor Reduce Leukocyte Adhesion and Platelet Aggregation in Sickle Cell Mice

Robert Figler PhD1,2, Michael Steele2, Marek Marcinkiewicz MD, PhD2, Renata Polanowska-Grabowska PhD2, Kori Wallace2, Susi Srinivasan PhD2, Lynn Hedrick PhD2,Klaus Ley MD2, and Joel Linden PhD2

1Adenosine Therapeutics, LLC, P.O. Box 4632, Charlottesville VA 22905 2Cardiovascular Research Center, University of Virginia, Charlottesville VA 22908 raf2z@virginia.edu

It has become increasingly appreciated that chronic inflammation plays an important role in the etiology of vasoocclusive crisis (VOC) and the long-term pathologic manifestations of sickle cell disease (SCD). Central to this concept is evidence that VOC is not merely a result of the passive obstruction of the vasculature by deformed erythrocytes, but an active process involving heterotypic interaction of erythrocytes, leukocytes, platelets and the endothelium. In studies using a transgenic murine model of SCD (NY1DD) we have demonstrated the following: 1) increased adhesion of a monocytic cell line to primary aortic endothelial cells cultured from NY1DD mice; 2) increased aggregation of platelets isolated from NY1DD mice; 3) increased formation of circulating leukocyte/platelet aggregates in the blood of NY1DD mice; and 4) increased adhesion of leukocytes to the endothelium of NY1DD mice in vivo.

This evidence of inflammation provides a rationale for the use of anti-inflammatory agents in the treatment of SCD and vaso-occlusive crisis. Adenosine has been demonstrated to act through the adenosine A2a receptor (A2aR) found on bone marrow-derived cells to inhibit inflammation. ATL146e is an A2a-selective agonist that has already been shown to be safe in man and is currently in Phase III clinical trials. We investigated the effects of ATL146e on indices of platelet and leukocyte activation in SCD.

We used an exteriorized cremaster muscle preparation to study the microcirculation in vivo as a means of assaying the effect of ATL146e on leukocyte interactions with the endothelium. At baseline, NY1DD mice have more than three times the number of adherent leukocytes/field when compared to congenic C57BL/6 mice. Treatment with ATL146e caused a pronounced reduction in leukocyte adhesion to wild type baseline levels when administered by either infusion (Alzet mini-pump, 10 ng/kg/min, overnight) or bolus injection (intraperitoneal, 5 2g/kg, 30 minutes prior to cremaster exteriorization).

Single particle counting was used to study platelet aggregation in vitro. At baseline, platelets isolated from NY1DD mice and subjected to sheer stress exhibit significantly increased aggregation as compared to control platelets. When NY1DD mice were pretreated in vivo with ATL146e (Alzet mini-pump, 10 ng/kg/min, O/N) platelet activation in platelet rich plasma prepared from these animals and exposed to ADP added in vitro was reduced to below control levels. Further, flow cytometric analysis of whole blood indicates that NY1DD mice exhibit an increased number of circulating leukocyte/platelet aggregates and that this number can be significantly reduced by pretreatment with ATL146e.

These data support the hypothesis that anti-inflammatory therapy will reduce the severity of vaso-occlusive crises and indicate that adenosine A2a agonists may be clinically useful for the treatment of vaso-occlusive crisis in SCD.

Allosteric enhancers of A1 Adenosine Receptors Catalyze Oxidation of Thiols

Mahendra D. Chordia,a Molly Zigler Karlinsey, a Heidi Figler,b Ray A. Olssonc and Joel Lindenb

a.Department of Chemistry and b. Cardiovascular Research Center, University of Virginia, Charlottesville, VA 22901 c. Suncoast Cardiovascular Laboratory, University of South Florida, Tampa, FL 33612 mdc3x@virginia.edu

Adenosine an endogenous ligand for four subtypes of G protein coupled adenosine receptors. Important physiological functions of Adenosine-A1 receptor activation have been described in many tissues including kidney, heart, adipose tissue and brain. Adenosine or synthetic adenosine analogs that occupy the orthosteric A1 receptor binding site have limited therapeutic potential due to cardiac and renal side effects and poor penetration into the brain. Allosteric enhancers (AE) of adenosine receptors have been described.1 These compounds decrease the dissociation kinetics of agonist radioligands from A1 receptors and facilitate adenosine signaling. Due to their lipophilic properties AEs may have clinical potential for producing analgesia, anti-seizure activity and sedation. Allosteric agents have little activity in the absence of adenosine and may selectively amplify adenosine signaling in hypoxic, ischemic or inflamed tissues that produce high levels of adenosine. Aminothiophenes, including PD 81,723 were the first compounds reported to have AE.2 We recently found that certain 2-aminothiazoles also have AE activity.3

We now report evidence suggesting that AEs act as thiol oxidants.. Reducing agents including DTT, N-acetylcysteine, glutathione and TCEP all block the activity of AE's to influence agonist radioligand binding. All compounds that are active as AEs can catalyze the oxidization of thiols such as glutathione to disulfides such as GSSH. The oxidizing power of AEs is well correlated with the receptor activity. In addition we discovered that AE's render A1 receptors resistant to solubilization in digitonin; certain disulfides have AE activity and H2O2 a strong oxidant has AE-like activity. Based on these observations we hypothesize that AEs catalyze oxidation of one or more Cys residues on the extracellular surface of A1 receptors to promote either intramolecular or intermolecular disulfide bonds. Agonist binding to the A1 receptor may be stabilized by and facilitate this oxidation. We hypothesize that formation of disulfides is responsible for stabilization of the ternary complex between ligand, receptors and G-proteins. This paper will discuss the detailed progress made towards unraveling mechanistic aspect of allosteric enhancer action on A1 adenosine receptors.

Alteration of adenosine receptors in patients with chronic obstructive pulmonary disease

Katia Varani,1* Gaetano Caramori,2* Fabrizio Vincenzi,1 Ian Adcock,4 Paolo Casolari,2 Edward Leung,3 Stephen MacLennan,3 Stefania Gessi,1 Silvana Morello,4 Peter J Barnes4, Kazuhiro Ito,4 Kian Fan Chung,4 Giorgio Cavallesco,5 Gianfranco Azzena,5 Alberto Papi,2* Pier Andrea Borea1*

1Department of Clinical and Experimental Medicine, Pharmacology Section, and 2Department of Clinical and Experimental Medicine, Respiratory Disease Section, and 5Department of Surgical Sciences, Thoracic Surgery Section, University of Ferrara; Ferrara, Italy; 3 King Pharmaceuticals Research & Development, Cary, North Carolina, USA; 4Airway Disease Section, National Heart & Lung Institute, Imperial College London, UK. (*these authors contributed equally to this work).

Chronic obstructive pulmonary disease (COPD) is the fourth leading cause of mortality worldwide. Adenosine acts through four distinct receptors to mediate pro- and anti-inflammatory effects. The primary aim of this study is to investigate the expression in peripheral lung parenchyma, the major site of airflow obstruction in COPD, using immunohistochemistry, radioligand binding and real time quantitative polymerase chain reaction. Adenosine receptors were analyzed in age-matched smokers with COPD (n = 14) and smokers with normal lung function (control group; n = 20). A1, A2a, A2b and A3 receptors were differentially expressed in peripheral lung parenchyma. The affinity of A1, A2a and A3 receptors was significantly decreased in COPD patients compared with control group [Kd(A1) = 3.15 ± 0.19* versus 1.70 ± 0.14 nM; Kd(A2a) = 7.88 ± 0.68* versus 1.87 ± 0.09 nM; Kd(A3) = 9.34 ± 0.27* versus 4.41 ± 0.25 nM; *p < 0.01] whereas their density was increased [Bmax(A1) = 53 ± 4* versus 32 ± 3 fmol/mg protein; Bmax(A2a) = 852 ± 50* versus 302 ± 12 fmol/mg protein; Bmax(A3) = 2078 ± 108* versus 770 T 34 fmol/mg protein; *p < 0.01]. The affinity of A2B receptors was not altered but the density was significantly decreased in COPD patients compared with the control group (Bmax = 66 ± 5* versus 189 ± 16 fmol/mg protein; *p < 0.01). A significant correlation was found between the affinity and density of the adenosine receptors and forced expiratory volume in one second (FEV1)/forced vital capacity (FVC) ratio, an established index of airflow obstruction. In conclusion, this is the first report showing the presence of adenosine receptors differentially expressed in lung parenchyma in COPD compared with control smokers. These novel findings strengthen the hypothesis of a potential role played by adenosine receptors in the pathogenesis of COPD.

Altered Distribution, Signalling And Function Of A1 AND A2a Adenosine Receptors In The Brain Of Wag/Rij Rats With Genetic Absence Epilepsy Before And After The Disease Appearance

1D'Auro M., 1D'Alimonte I., 2Citraro R., 1Di Iorio P., 1Giuliani P., 1Caciagli F., 2De Sarro G., 1Ciccarelli R.

1Department of Biomedical Sciences, University of Chieti-Pescara. Chieti. Italy. 2Department of Experimental and Clinical Medicine, “Magna Graecia” University. Catanzaro. Italy. mimmidauro@yahoo.com

Adenosine shows anticonvulsant properties in different types of epilepsy, but its role in absence seizures is still to be defined. Here, we investigated the distribution as well as the signalling pathways and function of A1 and A2a adenosine receptors in genetically absence epileptic Wistar Albino Glaxo/Rijswijk (WAG/Rij) rats compared with August Copenhagen Irish (ACI) rats, not prone to develop this kind of epilepsy. In WAG/Rij rats, the disease onset occurs after 2 months of age, due to generation of abnormal oscillatory rhythms within the thalamocortical circuitry, including the reticular (nRT), ventroposterolateral and ventroposteromedial thalamic nuclei as well as the frontoparietal somatosensory cortex. Our study, carried out in young (1,5 months) and adult (6 months) animals, i.e. before and after seizure occurrence, focussed on cerebral areas above mentioned in comparison with hippocampus, not involved in seizure triggering. Immunohistochemistry and Western Blot analyses showed a Communications 115 Springer lower density of A1 receptors in thalamic and cortical areas from young WAG/Rij rats, as well as in the only nRt in adult epileptic animals as compared with ACI rats of the same age. In contrast, the distribution of A2a receptors was uniformly decreased and increased in all brain areas of young and adult WAG/Rij rats, respectively, as compared with matched control. Accordingly, the stimulation of A1 and A2a receptors was ineffective on the modulation of cAMP formation and 50 mM K+-evoked [C14]-glutamate release in brain slices from young epileptic in comparison with control. Moreover, the A2a receptor stimulation with 2-(4-(-2-carboxyethyl)-phenylamino)- 5′-N-ethylcarboxamido- adenosine (CGS21680) increased either cAMP intracellular levels or K+-evoked labelled glutamate release to a greater extent in brain slices from adult WAG/Rij than ACI rats. On the contrary, the A1 receptor activation with 2-chloro-N6-cyclopentyladenosine (CCPA) inhibited forskolin-induced cAMP formation, but not K+-evoked labelled glutamate release, more remarkably in brain slices from adult WAG/Rij rat brains than from matched control. We also examined the modulation of the mitogen-activated protein kinase (MAPK) system, constituted by ERK1/2, p38 and JNK, because mostly p38 and JNK, when activated, are regarded as promoters of tissue stress and injury. Either CGS21680 or CCPA activated the entire MAPK system substantially in adult animals, more remarkably in brain slices from adult epileptic rats than from control. Thus, the reduced number of A1 and A2a receptors in WAG/Rij rat brain during development as well as the greater distribution of excitatory A2a receptors in adult epileptic rats, coupled to a significant alteration in signalling pathways and function linked to the activation of both receptors might deeply influence the pathogenesis of absence seizures, confirming some findings on the epileptogenic role of adenosine in such a disease1,2.

AMP Deaminase and AMP-Regulated Protein Kinase Interaction in Heart Cells

Ryszard T Smolenski 1,2), Ada HY Yuen 2), Iwona Rybakowska 1), Tomasz Borkowski 1), Ewa M Slominska 1), Krystian Kaletha 1), Magdi H Yacoub 2)

1) Department of Biochemistry, Medical University of Gdansk, Poland and 2) Heart Science Centre, Imperial College London, U.K. r.smolenski@ic.ac.uk

The C34T mutation of the AMP deaminase 1 gene (AMPD1) induces protective effect demonstrated clinically in patients with dilated cardiomyopathy, ischemic heart disease and in donors of hearts for transplantation. However, the mechanism of this effect has not been established. The product of the gene, AMP deaminase (AMPDA) 1 is responsible for catabolism of AMP and decrease of this activity will lead to elevation of AMP concentration. This in turn may activate dephosphorylation pathway of AMP degradation leading to formation of adenosine. Alternatively, AMP could activate AMP regulated protein kinase (AMPK) and related signalling cascade that is known to induce cytoprotective mechanisms. We have shown earlier that this mutation leads to about 50% decrease in cardiac AMPDA activity and that increase in adenosine production is relatively small. The aim of this study was to evaluate whether AMPDA could modulate the activity of AMPK.

AMPDA was isolated from human skeletal muscle using two step column chromatography on phosphocellulose. A new assay system has been developed for measurement of AMPK activity using HPLC with UV and mass detection that allows simultaneous monitoring of the substrate peptide (AMARA) phosphorylation and adenine nucleotide concentration. This assay was used to study the effect of isolated AMPDA added to in vitro assay of AMPK activity in human heart homogenates performed in the presence or absence of AMP.

Addition of AMPDA at 1 µg of protein reduced the activity of AMPK from 1.16 ± 0.06 to 0.43 ± 0.10 nmol/min/mg prot (n = 7, p < 0.001). Surprisingly, AMPDA reduced the activity of AMPK even in the absence of AMP from 0.72 ± 0.10 to 0.33 ± 0.09 nmol/min/mg prot (p < 0.05). This effect was dependent on the amoun of the AMPDA added.

We conclude that AMPDA activity may affect the activity AMPK in the cell. This interaction has been observed even without AMP added to the assay, possibly through competing with or deamination of AMP already bound to AMPK. Interaction of AMP deaminase with AMP-regulated protein kinase may have important implications for the regulation of cell function and in particular may be responsible for beneficial effect of AMP deaminase inhibition in the heart.

An intracellular redox and mercury-sensitive site in a purinergic P2X receptor.

C. Coddou, J.G. Lillo, P. Bull and J.P. Huidobro-Toro.

Centro de Regulacioń Celular y Patología, J.V. Luco, Instituto MIFAB, Departamento de Fisiología, Facultad de Ciencias Biológicas, Pontificia Universidad Católica, Santiago, Chile. ccoddou@puc.cl

P2Xs are ionotropic receptors differentially modulated by transition trace metals that bind to specific allostericsites. Extracellular histidines are essential for copper and zinc modulation in either P2X4 and P2X2 receptors, however these residues are not relevant for mercury-induced modulation in either receptor. Mercury has oppositte effects on the P2X2 and P2X4 receptors, it potentiates the former but inhibits the latter. To identify the mercury binding-site we used chimeric receptors containing the P2X4 ectodomain in the backbone of the P2X2 receptor. Chimeric P2X4/2 and native receptors were expressed in X.laevis oocytes, the ATP-induced currents were analized by two electrode voltage-clamp. In the P2X4/2a chimera, 10 µM Hg2+ increased 3.7 ± 0.8 fold the ATP-responses and displaced leftward the ATP concentration-response curve lowering the EC50 from 4.8 ± 0.6 µM to 2.4 ± 0.8 µM. (n = 8, p < 0.05). In the P2X4/2b chimera (a splice variant of the P2X2 receptor, lacking a 68 amino acid sequence in the C-terminal region), mercury was inactive. Copper inhibited and zinc potentiated the ATP-responses much alike the P2X4 receptor, confirming the extracellular nature of the copper and zinc allosteric sites. We mutated Cys-430 by alanine, one of the 68 amino acid sequence of the P2X2 receptor. In the C430A mutant the mercury-induced potentiation was significantly less than in the native P2X2 receptor, although a smaller potentiation remained. The C430A mutant, of the P2X4/2a chimera was completely mercury-insensitive, suggesting that Cys-430 is a target for mercury. We test the potential role of Cys-430 as a redox sensitive residue with hydrogen peroxide (H2O2). As mercury H2O2 potentiated the activity of the P2X2 receptor by 2.8 ± 0.3-fold with 0.3 mM H2O2 (n = 5); the P2X4/2a chimera was also potentiated by H2O2 (1 mM, 1.9 ± 0.2-fold). No potentiation was observed in the P2X2 C430 mutant nor in the P2X4/2b chimera. Chemical alkylation of the P2X2 receptor with the membrane-soluble methyl methanethiosulphonate MMTS also abolished the H2O2-potentiation. The P2X4 receptor was inhibited by 1 mM H2O2 by 50%, and this effect was prevented with the addition of 1 mM DTT, a reducing agent. In conclusion we identified intracellular Cys-430 as a key amino acid for the potentiation induced by mercury and hydrogen peroxide on the activity of the P2X2 receptor. This is the first report identifying an intracellular redox site in P2X receptor channels highlighting an additional role of P2X receptors as sensors of the cellular redox potential. Funded by FONDAP 13980001 grant. The Millenium Institute MIFAB contributed to fund the CRCP Center.

Analysis of the role of extracellular cysteine residues in the P2Y12-receptor

Irina Algaier, Ivar von Kügelgen

Department of Pharmacology, Universität of Bonn, Reuterstrasse 2b, 53113 Bonn, Germany ialgaier@uni-bonn.de

ADP activates human platelets through two G-protein coupled receptors, the P2Y1- as well as the P2Y12-receptor. Four extracellular cysteine residues of the P2Y12-receptor protein have been proposed to form disulfide bridges between the extracellular receptor domains and to represent the sites of action of the active metabolites of clopidogrel [1,2]. An important role of cysteine residues for receptor function has previously also been shown for the human P2Y1-receptor [3]. In order to study the role of the extracellular cysteine residues of the human P2Y12-receptor we replaced cysteine residues by alanine residues by site-directed mutagenesis of the respective DNA sequence. Wild type P2Y12-receptor and mutant receptors were stably expressed in 1321N1-astrocytoma cells. Immunofluorescence staining was used to demonstrate the expression of the mutant receptors. Changes in receptor function were assessed by measuring 2-methylthio-ADP (0.01 pM to 1 µM) induced inhibition of cellular cAMP levels. The addition of isoprenaline 10 nM elicited an increase in cellular cAMP levels (average increase by 55 ± 4 pM cAMP per well). In cells expressing wild type receptors, 2-methylthio-ADP caused a concentrationdependent inhibition of the cAMP production with an EC50 concentration of 0.9 nM and a maximal inhibition by 47%. In cells expressing recombinant receptors in which both the residue Cys17 in the N-terminus as well as the residue Cys270 in the third extracellular loop were replaced by alanine residues (Cys17Ala/Cys270Ala), 2-methylthio- ADP caused an inhibition in adenylate cyclase activity with a reduced maximal response (maximal inhibition by about 10%). The same was true for cells expressing mutant receptors in which Cys97 (located near the exofacial end of transmembrane region 3, TM3) as well as Cys175 (in the extracellular loop 2) were replaced by alanine residues (Cys97Ala/Cys175Ala; maximal inhibition by about 10%). In contrast, in cells expressing Cys17Ala/Cys97Ala-, Cys17Ala/Cys175Ala-, Cys97Ala/Cys270Ala- or Cys175Ala/Cys270Ala-mutant receptors any inhibitory action of 2-methylthio-ADP was lost. Additional experiments were performed to study the action of thiol agents at the P2Y12-receptor. These compounds have also been proposed to interact with extracellular cysteine residues. p-Chloromercuribenzene sulfonic acid (pCMBS; 0.1 to 10 µM) and N-ethylmaleimide (NEM; 0.3 to 10 µM) abolished or markedly reduced the responses to receptor stimulation by 2-methylthio-ADP with EC50 concentrations of 0.9 µM (pCMBS) and 10 µM (NEM), respectively. Furthermore, in experiments performed under reducing conditions (presence of ascorbic acid 40 µM) receptor-mediated inhibition of adenylate cyclase activity was almost lost. In conclusion, our present results demonstrate that Cys17 (N-terminus) and Cys270 (third extracellular loop) form one disulfide bridge and that Cys97 (exofacial end of transmembrane region3, TM3) and Cys175 (second extracellular loop) the second bridge. Furthermore, the data indicate that the presence of only one intact disulfide bridges reduces the maximal responses to activation by agonists and that thiol reagents block the receptor. Hence, our findings are compatible with the view that the active metabolites of clopidogrel or prasugrel interact with the extracellular cysteine residues of P2Y12-receptor.

Application of adenosine 5′-triphosphate (ATP) infusions in palliative home care: a multiccentre randomized clinical trial

S. Beijer, P.C. Dagnelie, N.E.G. Wijckmans

Department of Epidemiology, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands. epid.unimaas.nl


Palliative care in terminal cancer aims at alleviating suffering of patients. Complaints like progressive fatigue, deteriorating performance status, weight loss and decreasing functional abilities have a substantial impact on the quality of life of the patient and his/her environment. So far, few if any therapies are available to combat these devastating complaints. A randomized clinical trial in patients with advanced lung cancer showed favourable effects of regular ATP infusions on body weight, fat and fat-free mass, muscle strength, fatigue and quality of life, relative to untreated controls[1]. However, so far, no evidence is available about the effectiveness of ATP in other types of cancer, nor for the application of ATP infusions in a non-clinical setting. Based on this, we have initiated a study 1. To reproduce the results, which were previously observed in lung cancer patients, in patients with other types of advanced cancer, and 2. To test the feasibility and safety of regular ATP infusions at the patient's home.

Subjects and methods:

The study is targeted on cancer patients with a life expectancy of less than 6 months, who are (partly) mobile and suffer from at least one of the following three complaints: fatigue, weight loss, or anorexia (lack of appetite). Only patients are included with a life expectancy of <6 months, and for whom no systemic treatment is available: medical treatment options are restricted to palliative/supportive care. Patients are recruited through oncology departments of five participating hospitals, and family doctors. After eligible patients have signed informed consent, patients are stratified for tumour type and presence of weight loss (yes/no) and then randomly assigned to ATP or no ATP. The first ATP infusion is given at the day care centre of the participating hospitals, subsequent infusions are given at the patients’ home by an experienced and trained nurse of the infusion team of regional community care organizations. Outcome parameters are assessed every two weeks, until 8 weeks.

Preliminary results:

Preliminary results will be presented at the meeting. So far, we have enrolled and randomised 80 patients of a wide variety of tumour types. One of the problems encountered is a high rate of drop-out due to rapid deterioration of patients’ condition, which is difficult to foresee at the time of inclusion: approximately 75% of patients complete 4 weeks of follow-up, and 50% complete the full 8 week study period.

ATP, adenosine, and P1/P2 receptors in the control of neutrophil chemotaxis.

Yu Chen1, Ross Corriden1,2, Linda Yip1, Yoshiaki Inoue1, Cindy Cheung1, Paul A. Insel2,3, and Wolfgang G. Junger1

University of California San Diego, Depts. of 1Surgery/Trauma, 2Pharmacology, and 3Medicine, San Diego, California 92103.

As a key cell of the innate immune system, polymorphonuclear neutrophils (PMN) play an important role in inflammation and host defense. It is well known that PMN accumulate at sites of infection and inflammation where they kill bacteria, but these cells can also damage host tissues by releasing cytotoxic mediators. PMN possess complex mechanisms that allow them to travel to the source of chemotactic mediators released from bacteria or inflamed tissues. Although many aspects of PMN chemotaxis have been identified in recent years, it remains unclear how PMN amplify subtle differences in the concentration of chemoattractants within a gradient field.

Recent data from our laboratory indicate that chemoattractants such as the bacterial formyl peptide fMLP induce the release of cellular ATP from PMN at the leading edge that is closest to the source of fMLP. HPLC analysis revealed that PMN rapidly hydrolyze released ATP to adenosine. Using real-time RT-PCR analysis of all known P1 adenosine and P2 nucleotide receptor subtypes, we found that PMN and PMN-like HL-60 cells predominantly express the P2Y2 nucleotide and A2a and A3 adenosine receptor subtypes. Elimination of extracellular ATP or adenosine with apyrase or adenosine deaminase (ADA), respectively, dose-dependently blocked PMN activation and chemotaxis toward fMLP. Inhibition of P2 and A3 receptors significantly reduced chemotaxis, while agonists of these receptors, but not of A2a receptors, increased chemotaxis. Using fluorescent antibodies and HL-60 cells expressing P2Y2- and A3-EGFP fusion proteins, we found that cells migrating in a chemotactic gradient redistribute stimulatory A3 receptors to the leading edge (Fig.), while P2Y2 and A2a receptors remain uniformly distributed on the cell membrane. PMN of A3 knockout mice showed significantly reduced PMN migration in vitro and in a peritonitis model of inflammation.

Together these data imply that ATP release and positive feedback through P2Y2 receptors are essential steps needed for the amplification of initial chemotactic signals, while the accumulation of A3 receptors at the leading edge and positive feedback through adenosine predominantly generated at the leading edge maintains the locomotion of cells in a chemotactic gradient field. Based on these findings and conclusions, we propose a “pull-push” model of PMN chemotaxis whereby ATP release and adenosine formation together with stimulatory A3 receptors that are enriched at the leading edge increase cell migration, while suppressive A2a receptors at the receding edge facilitate membrane retraction.

This study was supported in part by grants from the National Institutes of Health (GM-51477, 60475, & 66232). Open image in new window

Fig. Confocal image of HL-60 cell expressing A3-EGFP receptor at the leading edge. The cell is migrating from the top right to the bottom left corner of the image shown.

ATP as a neurotrasmitter of pain in migraine: a functional role for P2Y receptors in primary cultures from mouse trigeminal sensory ganglia.

Fumagalli M.1‡, Ceruti S.1‡, Verderio C.2‡, Abbracchio M.P.1‡

1Laboratory of Molecular and Cellular Pharmacology of Purinergic Trasmission, Department of Pharmacological Sciences, University of Milan and 2 CNR Institute of Neuroscience, Cellular and Molecular Pharmacology, Department of Medical Pharmacology, Milan. marta.fumagalli@unimi.it All authors equally contributed to this work

Migraine is a common episodic headache disorder affecting more than 10% of the population (1). Despite the availability of adjuvant drugs, treatment of migraine remains unsuccessful in a significant number of patients, suggesting that yet-unidentified causative agents remain to be discovered. In analogy to its role in other chronic pain states, ATP could represent a potential algogenic factor in trigeminal sensory terminals during migrainous attacks (2). On this basis, the present study was undertaken to characterize, in primary cultured cells from mouse trigeminal ganglia, the presence and function of ATP/P2 receptors, with special focus on P2Y receptors. As a first step, we have characterized the percentage of the various cell types in our culture (i.e. neurons, astrocytes, and oligodendrocytes) by means of flow cytometry. Data showed that about 50% of cells were positive for oligodendrocyte markers, 20% for the astroglial marker GFAP, and 10% were neurons. Results were also confirmed by immunocytochemistry. Contaminating fibroblasts probably represents the remaining population. As a second step, we have characterized by RT-PCR the expression of all P2Y receptors cloned from rodent tissue (P2- Y1,2,4,6,12,13,14) and of some “P2Y like” orphan G-protein-coupled receptors (oGPCRs). Results showed the presence of all P2Y receptors, together with four oGPCRs. Single-cell calcium imaging of Fura-2 loaded cells was then used to dissect the functionality of specific receptor subtypes by measuring their coupling to intracellular calcium ([Ca2+]i) in the presence of subtype-selective P2 receptor agonist and antagonist ligands. We found that, despite the presence, at the mRNA level, of all P2Y receptors, only some of them are functionally coupled to increases of [Ca2+]i. In particular, application of ADP or 2MeSADP, agonists at P2Y1, P2Y12, P2Y13, induced a [Ca2+]i increases in approximately 20% of a sub-population of small and medium-size neurons and in almost all satellite-like glial cells. Treatment with the P2Y1 selective antagonist MRS2179 potently reduced [Ca2+]i responses induced by ATP in both types of cells. Furthermore, in satellite-like glial cells, the ADP-induced [Ca2+]i responses were significantly reduced by AR-C69931MX, an antagonist selective for P2Y12 and P2Y13receptors, suggesting that a component of the [Ca2+]i responses evoked by ADP derives from activation of P2Y12and P2Y13. We also observed [Ca2+]i increases in neurons and satellite-like glial cells stimulated with UTP, known to activate P2Y2 and P2Y4 receptors, and with UDP, which activates P2Y6 receptors. Thus, in mouse trigeminal ganglia, P2Y1, and, to a lesser extent, P2Y2,4,6,12,13 contribute to ATP-induced calcium-dependent signalling. The elucidation of the interaction of these receptors with ionophoric P2X receptors (in particular, P2X3) after sensitization of trigeminal neurons with algogenic stimuli (e.g., NGF, BDNF or bradykinin), may help identifying new potential targets for the development of novel antimigraine drugs.

Sponsored by Comitato Telethon #GGP04037 to MPA. AR-C69931MX was a kind gift of The Medicines Company, Waltham, MA, USA.

ATP inhibits the inflammatory response in stimulated whole blood, even under conditions of severe oxidative stress

Els L.R. Swennen1, 2, Aalt Bast2, Pieter C. Dagnelie1

1Department of Epidemiology, NUTRIM, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands.

2Department of Pharmacology and Toxicology, NUTRIM, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands. e.swennen@farmaco.unimaas.nl

Over the past decades, evidence has accumulated which indicates that extracellular nucleotides and nucleosides may be important regulators of inflammatory and immune responses. Many diseases in which inflammatory reactions are involved are associated with oxidative stress.

The purpose of the present study was to determine the effects of ATP on TNF-α, IL-10 and IL-6 release in stimulated whole blood in the absence and presence of oxidative stress. Blood samples were drawn from healthy volunteers and incubated with ATP and lipopolysaccharide + phytohemagglutinin for 24 h, without and with H2O2 as an inductor of oxidative stress.

In the absence of H2O2, ATP at 100 µM and 300 µM induced a reduction in TNF-α secretion by 32 ± 8% (mean ± SEM) and 65 ± 4%, respectively. Furthermore, these ATP concentrations induced an increase in IL-10 secretion by 93 ± 5% and 166 ± 7%1. ATP-γ-S and ADP also inhibited TNF-α release, but only ADP showed a stimulatory effect on IL-10 release. Co-treatment with adenosine deaminase did not reverse the ATP effect on TNF-α and IL-10. In the presence of 1, 5 and 10 mM H2O2, ATP at concentrations of 100 µM and 300 µM still inhibited TNF-α release and stimulated IL-10 release in stimulated blood{2.

Our results demonstrate that even under circumstances of severe oxidative stress, ATP has marked anti-inflammatory properties in stimulated whole blood.

ATP metabolism in human blood

Coolen JCM1,2, Swennen ELR1,2, Bast A2, Dagnelie PC1

1 Department of Epidemiology, NUTRIM, Maastricht University, Maastricht, The Netherlands

2 Department of Pharmacology and Toxicology, NUTRIM, Maastricht University, Maastricht, The Netherlands erik.coolen@epid.unimaas.nl


Cells release ATP into the extracellular environment in order to retain homeostasis and in response to mechanical and pathophysiological stimuli (e.g. ischemia, hypoxia). Plasma ATP measurement is complicated by 1) degradation of ATP by circulating and cell-based ectonucleotidases and by 2) ATP release from erythrocytes and platelets. The aim of this study was to investigate the degradation profile of ATP in plasma after addition of ATP to blood in the absence or presence of LPS/PHA stimulation, because our group showed earlier that ATP inhibits the inflammatory response induced by LPS and PHA in blood.


Fresh blood was aliquoted into 24-well plates and incubated with 300 µM ATP for 30 min at 37°C and 5% CO21. Samples obtained 0, 0.25, 0.5, 1, 2, 4, 8 and 24 hours after incubation with LPS/PHA or medium (control) were disposed of the cellular fraction and proteins by centrifugation. In a single run, ATP and metabolites were separated by methanol-gradient elution over a reversed-phase HPLC column, quantified by peak area through UV-absorption and identified by comparison of peak retention times with standards.


Results indicated continuous ATP degradation to ADP and AMP, which was complete by 4 hours. The level of ADP initially remained constant, but started to decline after 30 min. AMP increased during the first two hours, but slowly declined between 2 and 8 hours. Relatively high levels of hypoxanthine were observed after 24 hours. By that time, ATP, ADP and AMP were totally depleted. Similar results were obtained after incubation in the presence of LPS/PHA or medium.


The simultaneous decline of ATP and ADP after 30 min possibly contributed to the strong increase of AMP and hypoxanthine. The decline of AMP after 2 hours, which was probably caused by ongoing depletion of its precursors ATP and ADP, did not lead to an increase in inosine or adenosine levels. This may have been caused by uptake by erythrocytes and intracellular conversion to inosine and hypoxanthine, followed by secretion from the erythrocytes2. Finally, the increase in plasma hypoxanthine levels after 2 hours was not matched by its degradation products xanthine and uric acid, which would suggest conversion to other products.


Our results so far indicate a slower ATP breakdown in blood than anticipated, which is accompanied by a rise in hypoxanthine after 24 hours.

ATP/P2Y2 receptor-mediated intimate interaction between astrocytes and pericytes

Schuichi Koizumi1, Kayoko Fujishita1, Koji Sueishi1,2, Yasufumi Kataoka2, Satoko Ohkubo1 and Kazuhide Inoue3

1Division of Pharmacology, National Institute of Health Sciences, 1-18-1 Kamiyoga, Setagaya, Tokyo 158-8501, Japan.

2Department of Pharmaceutical Care and Health Sciences, Faculty of Pharmaceutical Sciences, Fukuoka University, 8-19-1 Nanakuma, Jonan, Fukuoka 814-0180, Japan.

3Graduate School of Pharmaceutical Sciences, Kyusyu University, 3-1-1 Maidashi, Higashi, Fukuoka 812-8582, Japan. skoizumi@nihs.go.jp

Brain pericytes are cells that are located in ablumenal site of small blood vessels or capillaries with patchy structure. Since capillaries lack smooth muscles and pericytes express α-smooth muscle actins, it is suggested that pericytes may control microcirculation by regulating tonus of capillaries. However, contraction or relaxation of brain pericytes is largely unknown. We previously reported that astrocytes release ATP that mediates glia-to-neuron communications as well as glia-to-glia communications. In addition to this, astrocytes form endfeet to surround the vascular wall, suggesting that astrocytic ATP may also control vascular functions. Exogenously applied ATP and its analogues produced rises in [Ca2+]i in cultured pericytes obtained from rat forebrain with pharmacological profile of P2Y2 and P2Y1 receptors. Interestingly, when stimulated, pericytes showed contractile responses, which was inhibited by the P2 receptor antagonists. In addition, pericytes released ATP in response to mechanical stimulation or even spontaneously, and produced elevations in [Ca2+]i, suggesting that pericytes could control function of capillaries by releasing endogenous ATP. We also demonstrate astrocytic control of pericytes or vascular tonus, which is also mediated by ATP/P2 receptors.

ATP reduces lactate dehydrogenase (LDH) and triacylglycerol levels in lung cancer patients- a potential explanation of ATP's favourable clinical effects?

P.C. Dagnelie, T. Brouwer

Department of Epidemiology, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Dagnelie@epid.unimaas.nl


Cancer-associated cachexia is a clinical syndrome of progressive weight loss associated with profound changes in host metabolism, including elevated lipolysis, proteolysis, and liver gluconeogenesis, and causing elevated plasma levels of lactate dehydrogenase (LDH) and triacylglycerols (TAG). At the same time, elevated LDH levels are a well-known marker of tumour progression and reduced survival. We previously reported that ATP infusions had marked beneficial effects on weight and muscle mass in patients with advanced lung cancer,1 as well as on survival in a subgroup of cachectic patients.2


To assess the effect of ATP infusion on plasma LDH and TAG levels in patients with advanced lung cancer.


Blood values from our previously published clinical trial in patients with non-small-cell lung cancer (NSCLC, N = 58, stage IIIB/IV) were analysed. Patients were randomly assigned to either supportive care (control), or ATP infusions (total of 10 courses of 30 h, given at 2- to 4-week intervals over 24 weeks). Differences over time were appraised by repeated-measures analysis.


At baseline, ATP and control groups were similar with regard to prognostic variables, including weight, tumour stage, and plasma values of LDH and TAG. Over the 24-wk study period, LDH and TAG increased in the control group, indicating progression of tumour growth and deterioration of the cancer cachexia syndrome. In contrast, LDH and TAG remained stable in the ATP group (between-group difference: P < 0.05 for both LDH and TAG).


The favourable effects of ATP on LDH and TAG levels in patients with advanced NSCLC may be due to anti-inflammatory effects of ATP. Swennen et al.3 recently showed that ATP inhibits the LPS-PHA inflammatory response in blood through inhibition of TNF-alpha release. TNF-alpha (a pro-inflammatory cytokine) has been associated with 1. elevated glucose-lactate cycling and 2. decreased lipoprotein lipase activity. Suppression of TNF by ATP could result in down-regulation of LDH production and increased clearance of plasma TAG, explaining the observed metabolic effects of ATP.


Our study demonstrates that ATP infusion has favourable effects on both serum LDH and TAG concentrations. These effects may partly explain the beneficial effects of ATP on body weight and survival in patients with NSCLC. Further investigations are needed for better understanding of the underlying mechanisms of these effects.

ATP Release Through Connexins Hemichannels in Xenopus Laevis Oocytes

Laia Bahima1, Xénia Grandes1, Jordi Aleu1, Mireia Martín-Satué1, Joan Blasi1, Jordi Marsal1, Luis C. Barrio2, Bulat Ziganhin3 and Carles Solsona1

1Laboratory of Molecular and Cellular Neurobiology, Department of Pathology and Experimental Therapeutics, IDIBELL-Medical School, University of Barcelona, Bellvitge Campus, Feixa Llarga s/n, E-08907 L'Hospitalet de Llobregat, Spain.

2Unit of Experimental Neurology, Research Department, “Ramón y Cajal” Hospital, Carretera de Colmenar Viej km 9, 28034 Madrid, Spain.

3Department of Pharmacology, Pharmacognozy and Botany, Kazan State Medical University, 49 Butlerov Street, Kazan 420012, Russia solsona@ub.edu

Usually, the controlled secretion of ATP occurs through the exocytosis of granules and vesicles. However, in some cells, and under certain circumstances, other mechanisms control ATP release. In gap junctions, connexins (Cx) link the cytoplasm of two adjacent cells by establishing an intercellular channel and control the passage of ions and molecules up to 1 kDa. The channel is formed by two moieties called hemichannels or connexons. It has been suggested that connexons may be an alternative pathway for ATP release. Connexins, like Connexin 38 (Xenopus endogenous connexin) and Connexin 32 (Cx32) can, under certain circumstances, permeate for ATP (Bahima et al., 2006). We have investigated ATP release through hemichannels formed by Cx32. This protein is expressed in most of human organs, but, particulary, Cx32 mutations present in Schwann cells produce X-linked Charcot Marie Tooth disease (CMTX). Expressing Cx32 in experimental model of Xenopus oocytes, we have combined the use of two electrode voltage clamp to record the ionic current generated by the hemichannels and the bioluminscent reaction of luciferin-luciferase. The hemichannels formed by Cx32 were estimulated by depolarizing pulses resulting in an outward current. After this estimulation, during the repolarization period, there was an inward tail current coincident with the release of ATP from the oocyte. The area of this tail current correlated with the amount of ATP released. Moreover, when we treated the Cx32 injected oocytes with 5 µg/ml Brefeldin A, a vesicular assembly blocker, there were no significant changes in ATP release, indicating that there was no exocytotic process implicated on it. On the other hand, in oocytes coexpressing Cx32 and P2X1 receptor, we found a larger tail current because the autoactivation of the oocyte, due to the opening of P2X1 receptor by the ATP released. Altogether these results strongly relate the release of ATP to the activation of Cx32 hemichannels in Xenopus oocytes.

Acknowledgements: This work is supported by grants from the Spanish Government, MCYT (SAF2005-00736/BFU2005-02202).

ATP Stimulates Reactive Oxygen Species Generation by NADPH Oxidase via P2X7 Receptors

Hewinson, J., Ward, S.G., & MacKenzie, A.

Department of Pharmacy and Pharmacology, University of Bath, Bath, UK, BA2 7AY. prsjh@bath.ac.uk

Reactive oxygen species (ROS) play diverse roles in host defence and inflammation. They have important functions in pathogen destruction, as highlighted in diseases such as chronic granulomatous disease, and have an emerging role in the inflammatory process, especially in the regulation of cell adhesion molecule expression. A few publications have shown that ROS generation in immune cells may be stimulated by ATP1,2. However the receptors and pathways involved have not be defined in any detail, but are important in understanding the role of ATP in inflammation and disease. In this work ATP-induced ROS generation in the human macrophage was investigated1,2. However the receptors and pathways involved have not be defined in any detail, but are important in understanding the role of ATP in inflammation and disease. In this work ATP-induced ROS generation in the human macrophage was investigated

All studies were performed using the human monocyte cell line THP-1 differentiated with phorbol 12-myristate 13-acetate and lipopolysaccahride, and real-time ROS generation was measured in cells pre-loaded with the ROS-sensitive fluorescent compound dichlorofluorescin diacetate. In these cells ATP induced ROS generation was dose-dependent with a peak occurring at 3–5 mM (EC50 269 µM). ROS generation was not completely inhibited using 1-[N,O-bis-(5-Isoquinolinesulfonyl)-N-methyl-L-tyrosyl]-4-phenylpiperazine (KN-62; 10 µM gave 52% inhibition with 3 mM ATP) which, taken with the dose response data, suggests P2 receptors other than P2X7 may also signal ROS production. The detection of ATP (3 mM)-induced ROS generation was inhibited using N-acetylcysteine (an antioxidant, concentrations ≥1 mM gave complete inhibition) and diphenyleneiodonium (concentrations ≥30 µM gave complete inhibition) but not with allopurinol (≤30 µM) or Nω-Nitro-L-arginine methyl ester (≤100 µM). Therefore NADPH oxidase and not xanthine oxidoreductase (XOR) or nitric oxide synthase is responsible for the generation of ROS, even though the cell lysates were shown to contain active XOR. None of the effective compounds inhibited ATP-induced calcium influx into Fluo-4 loaded cells therefore discounting the possibility that direct inhibition of P2 receptors was occurring. In calcium-free medium ATP-induced ROS generation was completely inhibited and pre-treatment of cells with 10 µM LY294002, 100 nM wortmannin, and 10 µM PD98059 resulted in partial inhibition of this pathway (48, 29, and 29% inhibition, respectively), whereas treatment with 10 µM Y-27632 (a ROCK inhibitor) and 0.3% butan-1-ol (a phosphatidic acid formation inhibitor) had no effect. Therefore these results suggest the involvement of phosphatidyl inositol-3 kinase (PI3K) and extracellular-signal regulated kinase (ERK) in the ATP-induced ROS generation signalling pathway. All effective compounds were tested for cytotoxicity using a lactate dehydrogenase assay. At the concentrations tested none of the compounds were cytotoxic.

These results suggest that ATP can signal the upregulation of ROS production in human macrophages. The generation of ROS is ATP dose-dependent, occurs by NADPH activation, and involves a calcium-dependent pathway that may rely on PI3K and constitutes a pathway that potentially has a role in host-defence and inflammation.

ATP-consuming and ATP-generating enzymes secreted by pancreas

Gennady Yegutkin1, Sirpa Jalkanen1 and Ivana Novak2

1MediCity Research Laboratory and Department of Medical Microbiology, Turku University and National Public Health Institute, Turku, Finland; 2August Krogh Building, Institute of Molecular Biology and Physiology, University of Copenhagen, Copenhagen, Denmark. gennady.yegutkin@utu.fi

In exocrine glands, ATP and other nucleotides/sides are thought to be important regulators of salt and fluid transport. In rat pancreas ATP is released from acini into the series of excurrent ducts that possess P2 receptors and also adenosine receptors, as shown by the recent study (1). Close to acini the ATP concentrations are in the high micromolar range. However, in final pancreatic juice collected from the main duct of stimulated pancreas we detected very low concentrations of ATP and high nucleotidase activity (2). Thus, in order to understand acinoductal paracrine regulation, it is important to determine which ecto-enzymes are present in pancreatic secretion. For this purpose, in vivo experiments on rat pancreas were performed and the major purine-converting exchange enzyme activities were determined in pancreatic juice.

Animals were stimulated with cholecystokinin (CCK-8) infusion (5.6 pmol/min/per 200 g animal) and pancreatic juice was collected from the common pancreatic bile duct. Purine-converting activities and transfer of γ-phosphate by pancreatic juice were determined by incubation of samples with [14C], [3H] or [γ−32P] labelled nucleotides, with/without cold nucleotides as second substrates, followed by subsequent separation by TLC and autoradiography (3). Pancreatic juice for enzymatic analysis was collected after 60 min of continuous CCK-8 stimulation when secretion was about 3 µl/min. The time-courses of [14C]ATP and [3H]ADP hydrolyses show that pancreatic juice contains enzymes that convert ATP to ADP, which is further rapidly converted to AMP. Data indicate that the juice contain NTPDase that can hydrolyze both ATP and ADP about equally well, i.e. CD39. Reverse-phase HPLC analysis additionally shows that this enzyme has broad substrate specificity towards other nucleotides, UTP and ITP. In addition, secretion contains ecto-5′nucleotidase, i.e. CD73, converting AMP to adenosine. Activities of non-specific phosphatases, ecto-nucleotide pyrophosphatase/phosphodiesterases and adenosine deaminase were negligible. In addition to hydrolytic enzymes, there were also counteracting ATP-generating enzymes in pancreatic juice, adenylate kinase and NDP kinase, capable of sequentially phosphorylating AMP via ADP to ATP.

Taken together, following CCK-8 stimulation of pancreas, pancreatic juice contains significant activities of enzymes consistent with CD39 and CD73. There are also enzymes with lower activities that are able to regenerate ATP from AMP and ADP, where other NTP can be used as donors of γ-phosphoryl groups. This newly discovered richness of secreted enzymes underscores the importance of purinergic receptors in signalling along pancreatic ducts lumen, and regulated release of ATP-hydrolyzing and ATP-generating enzymes. The projects were supported by the Danish Medical and Science Research Councils

Attenuation of Reperfusion Lung Injury and Apoptosis by A2A Adenosine Receptor Activation is Associated with Modulation of Bcl-2 and Bax Expression and Activation of Extracellular Signal-Regulated Kinases

Julia Rivo,1 Evelyne Zeira,2 Eithan Galun,3 Sharon Einav,4 Joel Linden,5Idit Matot6

1,4,6Department of Anesthesiology & Critical Care Medicine, 2,3 Goldyne Savad Institute of Gene Therapy, Hadassah University Medical Center, The Hebrew University of Jerusalem, Jerusalem, Israel, 5Departments of Medicine and Pharmacology, University of Virginia, Charlottesville, Virginia, USA. iditm@hadssah.org.il


Adenosine receptors (AR) and extracellular signal-regulated kinases (ERK) have been implicated in tissue protection and apoptosis regulation during ischemia-reperfusion (IR) injury. This study tests the hypothesis that reduction of reperfusion lung injury following A2AAR activation is associated with attenuation of apoptosis, modulation of ERK activation, and alterations in anti- and pro-apoptotic protein expression (Bcl-2 and Bax, respectively).


The arterial branch of the left lower lung lobe in intact-chest, spontaneously breathing cats was occluded for 2 hr and reperfused for 3 hr (IR group). Animals were treated with the selective A2AAR agonist ATL313 given 5 min before reperfusion alone, or in combination with the selective A2AAR antagonist ZM241385.


Western blot analysis showed significant reduction in expression of Bcl-2 and increase in expression of Bax after reperfusion compared with control lungs. Phosphorylated ERK1/2 levels were also increased following reperfusion. Apoptosis was preferentially increased in alveolar epithelial cells. Compared to the IR group, ATL313 markedly (P < 0.01) attenuated indices of injury and apoptosis including the percentage of injured alveoli, wet/dry weight ratio, myeloperoxidase activity, in situ terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) positive cells, and caspase 3 activity and expression. Furthermore, compared with reperfused lungs, in ATL313-pretreated lungs, Western blot analysis demonstrated substantial ERK1/2 activation, increased expression of Bcl-2, and attenuated expression of Bax. The protective effects of ATL313 were blocked by pretreatment with ZM241385.


In vivo activation of A2AAR confers protection against reperfusion lung injury. This protection is associated with decreased apoptosis and involves ERK1/2 activation and alterations in anti-apoptotic Bcl-2 and pro-apoptotic Bax proteins. Open image in new window Western blot analysis of Bcl-2 and Bax protein levels in control, nonsichemic group (I), ischemia reperfusion (IR) {II}. ATL313 continuous infusion of 1 ng · kg1 · min1 administered before IR (III), ATL313 continuous infusion of 10 ng · kg1 · min1 administered before IR (IV). ZM241385 (V). (a) Representative immunoblots (b) Densitometric scan of Bcl-2 and Bax in lung tissue from the different groups. The intensities of bands in control treated tissue samples (no ischemia or reperfusion, group I) were expressed as 100%. Data (mean ± SEM) were obtained from Western blot assays performed at the end of reperfusion. Immunodetection of actin was performed as an internal control. *p<0.05 compared with groups I, III, IV. **p<0.01 compared with groups II, V.

Augmentation of cutaneous hypersensitivity by perceived acute stress is mediated by the adenosine A1 receptor.

Sneha Mathew, Tuere Wilder, Bruce N Cronstein, Stephen J Oliver

Department of Medicine, Division of Clinical Pharmacology, New York University School of Medicine, New York, NY, USA. olives03@med.nyu.edu

Adenosine is utilized by the body's tissues for modulation of a diverse range of activities that include sleep, cognitive function, immunity and pain perception. Acting through its specific receptors, adenosine modulates neurotransmitter function and release, behavior, systemic immunity and cell metabolism. Adenosine is generated extracellularly by degradation of released adenine nucleotides (ATP, ADP, AMP) and is constitutively present at low levels in all tissues. However, extracellular adenosine levels can markedly increase under metabolically stressful conditions such as inflammation or hypoxia. Biological systems have evolved to utilize extracellular adenosine levels to activate protective responses against physiological stressors such as ischemia and infection. Acting through the CNS, perceived psychological stress also affects biological functions within the body. In particular, differential effects of acute and chronic stress on skin immunity have been demonstrated in the well-established rodent model of non-painful restraint stress; with acute stress enhancing and chronic stress suppressing cutaneous delayed-type hypersensitivity (DTH), an antigen-specific immune response known to be mediated by CD8+ T cells1, 2. In this study, we utilize the restraint stress model and the hapten dinitrofluorobezene (DNFB, 1% in 4:1 acetone/olive oil) to induce delayed type hypersensitivity in the skin. In WT mice undergoing restraint stress DTH was augmented at 48 hrs (mean change in pinna thickness, non-stressed [NS] vs acute stressed [AS]: 0.08 mm vs 0.31 mm, respectively, p = 0.008). In contrast, stress does not enhance DTH responses in A1KO mice (mean change NS vs AS, 0.18 mm vs 0.19 mm, respectively, p = 0.73). Further studies have failed to show any differences in stress-augmented plasma corticosterone and urinary norepinephrine levels between A1 receptor knockout (KO) and WT mice. Also, both A1 receptor KO and WT mice respond similarly to acute stress with rapid corticosterone-mediated deployment of monocytes and lymphocytes out of the blood vasculature into the surrounding tissues. However, acute stress associated increases in circulating blood neutrophils, known to be mediated by catecholamine release, were not seen in A1 KO mice (mean cell count, NS vs AS: 1.54 × 103 cells/µl vs 1.62 × 103 cells/µl, p = 1.000), in contrast to their WT littermates (mean cell count, NS vs AS: 1.18 × 103 cells/µl vs 4.18 × 103 cells/µl, p = 0.009). Others have suggested that the increased antigen specific T cell responses occurring with acute stress are due to norepinephrine-driven dermal dendritic cell (DC) migration to draining lymph nodes2. Using FITC (1% in 1:1 acetone/dibutylphthalate) topically applied to shaved ventral skin, we show that the enhanced rates of MHC class IIhigh CD11c+ DC migration to draining lymph nodes observed in acutely stressed WT mice (DCs containing FITC, NS vs AS, at 6 hrs, 17% vs 26.8%; at 24 hrs, 29.1% vs 39.8%) do not occur in the adenosine A1 receptor KO mice under similar conditions (NS vs AS, at 6 hrs, 27.4% vs 28.5%; at 24 hrs, 42.9% vs 38.1%). Thus, our results suggest that acute stress enhances cutaneous immunity and that adenosine receptor A1 signaling plays a critical role in acute stress-altered cutaneous DTH immunity, at least in part through its modulating effects on norepinephrine release and dermal DC migration rates.

Autocorrelation of Molecular Electrostatic Potential Surface Properties combined with Partial Least Squares Analysis as New Strategy for Predicting the Activity of Human A3 Adenosine Receptor Antagonists tool to generate ligand-based 3D-QSARs.

Magdalena Bacilieri1, Stefano Moro1, Barbara Cacciari2, Giampiero Spalluto3,

1 Molecular Modeling Section, Dipartimento di Scienze Farmaceutiche, Universitá di Padova, Via Marzolo 5, I-35131 Padova, Italy; 2 Dipartimento di Scienze Farmaceutiche, Universitá degli Studi di Ferrara, Via Fossato di Mortara 17-19, I-44100 Ferrara, Italy; 3 Dipartimento di Scienze Farmaceutiche, Universitá degli Studi di Trieste, Piazzale Europa 1, I-34127 Trieste, Italy magdalena.bacilierri@unipd.it

Ligand based 3D-structure-activity relationship (3D-QSAR) is an ensemble of strategies to study new compounds starting from molecule structures and their corresponding chemical properties. Throughout the different approaches, the combination of autocorrelation vectors1 of molecular electrostatic potential{2,3 (MEP) surface properties with the conventional partial least squares (PLS) analysis has been used for the prediction of the human A3 receptor antagonist activities. The autocorrelation allows in fact to compare molecules (and their properties) with different chemical structures and with different spatial orientation without any previous alignment (which is often not even possible).

A dataset of 358 structurally diverse human A3 receptor antagonists has been used to derive a novel ligand-based 3D-QSAR model. The molecular collection includes the most important chemical classes of human A3 antagonist structure-activity relationship. A remarkable quantitative model was generated, with a cross-validated correlation coefficient of 0.81 and a test set correlation coefficient of 0.82.4

The aim of this analysis was to employ a rigorously validated QSAR model for a possible application of virtual screening on available chemical databases to find out new compounds with high predicted activity. Open image in new window

BDNF-induced modulation of synaptic transmission requires functional adenosine A2A receptors in the mouse hippocampus.

Martire A, Domenici MR, Potenza RL, Pepponi R, Felici F, Tebano MT, Popoli P.

Istituto Superiore di Sanitá, Rome, Italy. patrizia@popoli@iss.it

Brain-derived neurotrophic factor (BDNF), a member of neurotrophin family, is able to enhance synaptic transmission and to induce long-term potentiation (LTP) either in vitro or in vivo. BDNF and its tyrosine kinase receptor (TrkB) are highly expressed in the hippocampus where an interaction with adenosine A2A receptors (A2ARs) has been recently demonstrated (Diogenes et al, 2004). The aim of the present work was to evaluate BDNF effects in the hippocampus of A2AR wild type (WT) and knock out (KO) mice (6 months of age). In electrophysiological experiments in hippocampal slices (450 um) from WT mice, application of BDNF (10–20 ng/ml, 30 min) increased the slope of excitatory post-synaptic field potentials (fEPSPs) recorded in the stratum pyramidale of CA1 region (113 ± 6.7%, P < 0.05 vs basal, N = 11). The co-application of the selective A2A antagonist ZM 241385 (50 nM) abolished the BDNF-induced potentiation of fEPSP slope (99.43 ± 1.05%, N = 6). Conversely, in slices from A2AR KO mice BDNF was no longer able to increase the fEPSP slope and it even reduced it (88.83 ± 3.29 P < 0.05 vs basal and vs WT, N = 12). Finally, enzyme immunoassay studies showed a tendency to a reduction in hippocampal BDNF levels in A2AR KO vs WT mice. These results suggest that in the hippocampus the presence and the tonic activation of A2ARs are necessary to allow BDNF-induced potentiation of synaptic transmission. Given the key role exerted by hippocampal BDNF on both synaptic plasticity and neuroprotection, it is of paramount importance to understand how the tone existing at A2ARs may influence the BDNF effects under both physiological and pathological conditions.

Benzo [d]isothiazol-3-yl-benzamidines: a new class of protective agents on culture of human chondrocytes stimulated by IL-1β

AM. Panico1, P. Vicini2, M. Incerti2, F. Garufi1, S. Ronsisvalle1.

1Dept. of Pharmaceutical Sciences, University of Catania, V.le Doria 6, 95125 Catania (Italy)

2Pharmaceutical Dept., University of Parma, Parco Area delle Scienze 27/A, 43100 Parma (Italy) panico@unict.it

Several articular pathologies evolve from a local inflammatory disease to a chronic process with distinct inflammatory and destructive components. The progressive damage of the articular cartilaginous tissue represents a central moment of the pathogenetic course, in which depletion of aggrecan, disregulated synthesis of matrix components like proteoglycans and collagen play a fundamental role [1]. Moreover, the functional alteration of cartilage results also from the interaction of different mediators, such as cytokines, e.g. interleukin-1β ?IL-1β), that induces high levels of nitric oxide (NO), inhibits collagen and proteoglycans synthesis, increases susceptibility to injury by other oxidants (e.g. superoxide anion and hydroxyl radicals, peroxides, and reactive oxygen species, ROS).

The culture of human chondrocytes stimulate by IL-1β has been established, as a profit experimental model, for reproducing the mechanisms involved in the pathophysiology of arthritic diseases.

In the present study we determined the in vitro effects on human articular chondrocytes for a series of N-benzo[d]isothiazol-3-yl-benzamidines hydrochlorides designed to improve the effectiveness of N-benzo[d]isothiazol-3-yl-amidines previously described by us as promising chondroprotective agents [2]. Open image in new window

We investigated the influence of the new amidines on the production of NO, MMP-3, ROS, GAGs, key molecules involved in matrix destruction.

Our results clearly demonstrate that N-benzo[d]isothiazol-3-yl-benzamidines properly substituted can block multiple cartilage destruction during the arthritic inflammatory process, as simulated in our experimental model.

Beyond Purinergic Receptors: Drugging the Rest of the Purine-Binding Proteome

S. Hall, A. Barabasz, T. Barta, J. Daw, L. Dubois, J. Eaves, P. Fadden, B. Foley, T. Freed, L. Geng, G. Hanson, K. Hardemann, L. Hinkley, M. Jenks, M. Hu, K. Huang, M. Lewis, L. Liu, W. Ma, J. Otto, J. Partridge, B. Pronk, J. Rice, A. Scott, M. Silinski, P. Steed, J. Veal, K. Verleysen, R. Ware, T. Wadkins

Serenex, Inc. 323 Foster Street, Durham, NC, 27701 USA shall@serenex.com

Purinergic receptors represent an attractive class of druggable targets within the purine-binding proteome, yet there are many other enzyme families of interest as therapeutic targets within this large superfamily. We estimate that the purine-binding proteome contains at least 2000 members, with kinases composing the largest fraction(25%) of this group. We report here a novel approach to screen hundreds of these targets in parallel, thus providing an efficient way to probe compound libraries as ligands to a broad variety of enzyme classes. We constructed and screened a targeted library of 8000 compounds using a novel affinity displacement assay (referred to as proteome mining). Multiple tissues were used as the source of native purine-binding proteins to profile this library. Hits were obtained for nearly 100 targets, including well validated targets such as dihydrofolate reductase and multiple kinases (e.g. CDK2, Her2, PDK1, AurA), and less well-validated targets such as the kinases Fer and Nek9. Importantly, the method also provided affinity information on potential toxicity targets (e.g. phosphorylase kinase, pyruvate carboxylase, glucose-6-phosphate dehydrogenase). Thus, this approach provides primary target affinity information as well as selectivity data across the entire purine-binding proteome. In addition, affinities of compounds for their protein targets could be determined directly from the same assay and these measurements were sufficiently quantitative and reproducible to drive the development of structure-activity relationships. Unexpectedly, selectivity within a gene family was not predictive of selectivity between gene families across the purine-binding proteome. The application of this screening approach to four diverse targets, dihydrofolate reductase, Hsp90, quinone reductase 2, and PI3kinase will be described. In each case, the use of the affinity displacement assay enabled the identification of novel, selective, orally active compounds. Taken together, these results support the use of proteome mining as both an alternative to single target biochemical screens and as a tool to drive SAR development. Open image in new window

Bioinformatic analysis of the P2X receptor family

Digby HR1 Sutcliffe MJ2 & Evans RJ1

1Department Cell Physiology and Pharmacology, University of Leicester

2School of Chemical Engineering and Analytical Science, University of Manchester hrd1@le.ac.uk

P2X receptors for ATP are a family of ligand-gated ion channels consisting of seven receptor subunits that associate most likely as homo- and heterotrimeric channels. They are expressed throughout the body and have a variety of physiological roles which means the development of P2X receptor subtype selective drugs could have wide ranging therapeutic benefits. For example, P2X1 receptors have been implicated in the regulation of thrombosis and P2X3 receptor antagonists may act as analgesics and regulate bladder function. There are no crystal structures available for the P2X receptor family, they have no obvious similarity to other ion channels or ATP binding proteins and motifs such as the Walker motif are not present as in other ATP-sensitive proteins. For this reason, mutagenesis studies have been relied upon to identify important residues involved in ATP action at the receptors. However, the need for a structural model is imperative to functional studies aiding rational drug design and the development of pharmacophores. A portion of the extracellular loop of P2X receptor proteins shares a degree of homology with the catalytic domains of class II aminoacyl-tRNA synthetases and a homology model of the rat P2X4 receptor has been published based on this similarity. However, existing models do not consider the entire protein sequence of the receptor and the likely trimeric structure. I have produced possible 3D models for the structure of the human P2X1 receptor using ATP-binding proteins with known structures and similar ATP binding environments to the P2X1 receptor as predicted by mutagenesis studies. The structural models attempt to represent the extracellular ligand-binding loop. The models are currently being tested by site-directed mutagenesis. Supported by the Wellcome Trust

Biological Significance of the Endogenously Expressed Truncated P2X7 Receptor Variant (P2X7−j) in Human Uterine and Skin Cancers

Xin Li MD PhDa, Lingying Zhou MDa, Ying-Hong Feng MD PhDb, George Gorodeski MD PhDa,c

Departments of Reproductive Biologya, Physiology and Biophysicsc, and Oncologyc, CASE (Case Western Reserve) University, Cleveland, Ohio, USA (gig@cwru.edu); and Pharmacologyb, Uniformed Services University of the Health Sciences, Bethesda, Maryland.

The P2X7 system plays an important role in spontaneously occurring apoptosis and cell-number regulation in the human cervix (1). Recent studies from our lab revealed the presence of naturally occurring truncated variant of the human receptor P2X7 in human cancer cervical epithelial cells. The novel protein (P2X7−j), a polypeptide of 258 amino acids, lacks the entire intracellular carboxy terminus, the second transmembrane domain, and the distal third of the extracellular loop of the full-length P2X7 receptor. Expression experiments in host cells that lack endogenously the P2X7 receptor revealed diminished ligand-binding and channel function capacities, and failure to form pores and mediate apoptosis in response to treatment with the P2X7 receptor agonist BzATP. The present study extended those preliminary experiments to other epithelial tissues, including the ectocervix (EctoCx), endometrium (EndoM), and skin. In the first experiment primary-secondary cultures of normal and cancer epithelial cells were generated on filters (2), and the degree of apoptosis was determined by the DNA solubilization assay (3) following 9 hrs of treatment with 100 2M BzATP. The results in Fig. 1 show diminished apoptosis in all three types of cancer cells compared to the corresponding normal cells. The second experiment determined mRNA (Fig. 2) and protein levels (Fig. 3) of the P2X7−j and Full-Length P2X7 receptor in normal and cancer ectocervix, endometrium, and skin tissues. mRNA levels were determined by Real-Time RT-PCR and normalized to GPDH; protein levels were determined by densitometry of the P2X7−j — specific 45–42 KDa bands off Western blots using Alomone-M0 anti P2X7 pAb (which recognizes the proximal extracellular loop of the P2X7 receptor) and normalized to tubulin. The data show that in the three types of tissues the ratios of P2X7−j/P2X7 mRNA (Fig. 2) and protein (Fig. 3) were greater in cancer than in normal cells. These results suggest a general pattern of upregulation of the P2X7−j or downregulation of the Full-Length P2X7 in epithelial cancers. Because the P2X7−j variant antagonizes the Full-Length P2X7 receptor (manuscript submitted), the present results suggest a novel pathophysiological mechanism of apoptosis inhibition through regulation of P2X7 protein function by its variant the P2X7−j. Dysregulated apoptosis could play a role in carcinogenesis. Open image in new window

Support: AHA-SDG-0030019N, NHLBI-HL65492 (YHF); NIH HD29924, AG15955, and an unrestricted grant by CytoCore Inc (GIG). Patent pending.

Bivalent adenosine antagonist-dopamine agonits for the treatment of Parkinson's disease

Aroa Soriano, Ruben Ventura, Vicent Casadó, Fernando Albericio*, Carmen Lluis, Miriam Royo*, Rafael Franco.

Molecular Neurobiology Unit. Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS). University of Barcelona. Diagonal 645. 08028 Barcelona and * Scientific Park PCB. University of Barcelona. Josep Samitier s/n. 08028 Barcelona. rfranco@ub.edu

G-protein-coupled (heptaspanning) membrane receptors are target for a wide range variety of diseases. Recent evidence shows that these receptors occur on the cell surface as homo and heterodimers. Therefore dimers but not monomers are the actual targets for therapeutical useful drugs (already marketed or under development). The heteromerization is one of the molecular basis involved in the negative cross-talk between adenosine and dopamine. The anti-parkinsonian actions of A2AR antagonists in humans (already in phase 3 clinical trials) are to a substantital degree caused by blocking the action of endogenous adenosine on A2AR of the A2AR/D2R heteromer. We have targeted A2AR/D2R heterodimers by the development of novel compounds that could potentially interact simultaneously with dopamine D2 and A2A receptors. Dopamine-adenosine bivalent ligands have mixed A2A receptor antagonist and D2 receptor agonist pharmacophores coupled by connecting spacers of variable lengths. These spacers are polyamide-poly (ethylene) glycol type oligomers composed by a precise number of monomer units [(NH-CH2CH2O-CH2CH2O-CH2CH2-NH-CO-CH2CH2-CO)] and were synthesized by a solidphase protocol. The potential of these bivalent ligands has already been confirmed in cell models and subsequently patented. Bivalent liagnds of given lengths are more potent than monovalent compounds and can simultaneously bind to both D2 and A2A receptors in the heterodimer. Their efficacy has been tested in cAMP level determination assays in transfected cells. A lead compound has been selected that has D2 receptor agonist activity and A2A receptor antagonists activity and higher relative efficacy than monovalent counterparts. These bivalent ligands prove that A2AR/D2R heteromers occur in striatum and they constitute novel therapeutical tools for PD. It is expected that these compounds are more efficacious than combined therapies using L-DOPA and adenosine antagonists and have less side effects.

Blockade of adenosine A2A receptors differently alters convulsive behaviour and prevents hippocampal damage in two models of temporal lobe epilepsy

Lisiane O. Porciúncula, Paula M. Canas, Catarina R. Oliveira, Rodrigo A. Cunha Ctr. Neuroscience Coimbra, Fac.Medicine, Univ.Coimbra, Portugal. (loporciuncula@yahoo.com)

Endogenous adenosine in the brain is thought to prevent the development and spread of seizures via tonic anticonvulsant A1 receptor-mediated effects (Fredholm et al., 2005). However, adenosine modulation results from a balanced activation of inhibitory A1 and facilitatory A2A receptors (A2ARs) and blockade of these A2ARs affords neuroprotection against different brain insults (Cunha, 2005). Thus, we tested the role of A2ARs on the development of behavioural seizures and hippocampal damage in the two models of epilepsy: kainic acid (KA) ip administration and amygdala kindling.

Male Wistar rats injected ip with KA (10 mg/kg, ip) displayed rapid (within 30 min) and severe convulsions (stage 4–5 of Racine's scale) and a pattern of neuronal damage (cresyl violet and FluoroJade staining) together with astrogliosis (GFAP immunoreactivity) and microgliosis (tomato lectin or CD11b immunoreactivity) in the hippocampus at 24 h or 7 days after KA injection. SCH58261 (0.05 mg/kg, ip, 30 min before KA) caused an inconsistent effect on convulsions (prevention, attenuation and no effect, 3 rats each) but abolished neuronal damage, astrogliosis and microgliosis in all 9 rats.

To confirm the role of A2ARs in the development of hippocampal damage resulting from behavioural seizures, we compared the effect of KA in wild type and A2AR knockout male C57Bl6 mice. As occurred in rats, KA (35 mg/kg, sc) injection in wild type mice triggered convulsions (stage 5–6 of modified Racine's scale) as well as hippocampal damage typified by neuronal damage, astrogliosis and microgliosis observed 24 h after KA injection. However, in A2AR knockout mice, KA injection caused a similar severity of convulsions but did not produce any hippocampal damage (no FluoroJade staining and no modification of GFAP- or CD11b-immunoreactivities compared to saline controls).

Finally, we investigated if A2AR blockade would interfere with epileptogenesis using a model of epilepsy (amygdala kinding) that causes an onset of seizure activity slower (in the range of days) compared to kainate injection (within 30 min). In vehicle injected control rats, tonic/clonic seizures were observed 8–10 days after starting the kindling protocol (sub-maximal stimulation 500 µA, 1 msec pulses at 50 Hz for 1 sec, twice a day, 10 AM and 4 PM). In contrast, in 90% of kindled rats, the administration of SCH58261 (0.05 mg/kg, ip, 30 min before each stimulation period) delayed and attenuated the severity of convulsions. Likewise, astrogliosis and microgliosis observed after 15 days were prevented by SCH58261.

These results provide the first evidence that endogenous adenosine participates in the development of epilepticinduced damage through A2AR activation. These A2ARs might also aggravate epileptogenesis but this only evident in a chronic (kindling) rather than acute (KA) model of epilepsy. This questions the role of adenosine as an anti-epileptic agent and suggests that A2AR antagonists may provide a novel strategy to control neuronal damage occurring in temporal lobe epileptic conditions. (Supported by FCT)

Blockade of adenosine A2A receptors prevents β-amyloid (Aβ1–42)-induced synaptotoxicity and memory impairment in rodents

Geanne M.A.Cunha1,2, Paula M.Canas1, Jiang-Fan Chen3, Catarina R. Oliveira1, R.A. Cunha1

1Ctr. Neuroscience Coimbra, Fac. Medicine, Univ. Coimbra, Portugal, 2Dept. Physiol. Pharmacol., Federal Univ. Ceara, Fortaleza, Brazil, 3Dept. Neurology, Boston Univ. School Medicine, MA, USA (canas.paula@gmail.com)

Alzheimer’s disease (AD) is the most common chronic neurodegenerative disease characterized clinically by an atrophy of hippocampal regions and a progressive cognitive impairment (Hardy and Selkoe, 2002). The parameters that correlate better with memory dysfunction in AD are the levels of soluble Aβ, mainly Aβ1–42, and a decreased density of nerve terminals in cortical areas (Hardy and Selkoe, 2002). Thus, a major lead for the development of novel therapeutic strategies for AD might be to explore mechanisms able to prevent this early synaptotoxicity caused by Aβ1–42.

Adenosine plays a prominent role in controlling brain neurodegeneration. In particular adenosine A2A receptor (A2AR) blockade confers a robust neuroprotection in chronic noxious brain conditions (Cunha, 2005). Accordingly, we showed that caffeine and selective A2AR antagonists prevent Aβ-induced toxicity in cultured neurons (Dall'Igna et al., 2003). Thus, we tested if the A2AR antagonist SCH58261 (0.05 mg/kg, ip daily for 15 days) prevented the synaptotoxicity and memory impairment observed in rats two weeks after administration of Aβ1–42 (2 nmol, icv). The rats displayed a mnemonic deficit (Y maze) and a decrease of synaptophysin immunoreactivity (evaluated by immunohistochemistry and Western blot analysis), but neither cell death (evaluated by cresyl violet staining), nor microglia recruitment (evaluated by CD11b immunohistochemistry) was observed. In contrast, in SCH58261-treated rats, Aβ1–42 failed to modify cognitive performance and there was no synaptotoxicity, i.e. reduction of synaptophysin immunoreactivity compared to control.

We attempted to confirm this key role of A2ARs in the development of the Aβ1–42-induced synaptotoxicity and memory impairment using a genetic deletion rather than a pharmacological blockade of A2ARs, i.e. using A2AR knockout mice. As occurred in rats, there was a mnemonic deficit and a decrease of synaptophysin immunoreactivity, but neither cell death nor microglia recruitment in wild type C57Bl6 mice two weeks after administration of Aβ1–42 (2 nmol, icv). In contrast, in preliminary experiments, none of these modifications were found two weeks after administration of Aβ1–42 (2 nmol, icv) in A2AR knockout mice.

These results show that the blockade of A2ARs, which are enriched in hippocampal synapses (Rebola et al., 2005), prevents Aβ-induced synaptotoxicity and consequent amnesia in rodents. This key role of A2ARs in the development of synaptic loss and memory impairment in an animal model of AD provides a rationale for a previous suggestion that caffeine consumption (an adenosine receptor antagonist) might attenuate the incidence of AD (Maia and de Mendonça, 2002). (Supported by CNPq Brazil, FCT, Pfizer award from SPN)

Blockade of P2Y1 receptors prevents the synaptotoxicity and memory impairment caused by Aβ1–42 administration

Ricardo J. Rodrigues1, Lisiane O. Porciúncula1, Paula M. Canas1, Christian Gachet2, Catarina R. Oliveira1, Rodrigo A. Cunha.

1Center for Neuroscience and Cell Biology of Coimbra, Inst. Biochemistry, Fac. Medicine, Univ. of Coimbra, Coimbra, Portugal.

2INSERM U.311, Etablissement Français du Sang-Alsace, 10 rue Spielmann, B.P.N°36, 67065, Strasbourg Cedex France rodriguesrj@clix.pt

Amyloid beta-peptides (Aβ) are accepted to be a major cause of neuronal death in Alzheimer's disease (AD) (Hardy and Selkoe, 2002) which is predated by a synaptic loss (Selkoe, 2002). Thus, a major lead for the development of novel therapeutic strategies for AD might be to explore mechanisms able to prevent this early synaptotoxicity caused by Aβ1–42. Upon cell damage there is an increase of the extracellular levels of ATP, which can activate P2 receptors (P2Rs). Brain possesses both ionotropic P2XR and metabotropic P2YR which can control neuronal damage (Volonte et al., 2003). Since we showed that P2X1–3Rs and P2Y1,2,4Rs are located in nerve terminals (Rodrigues et al., 2005), we now investigated if P2 receptors could control damage in cultured hippocampal neurons and, in particular, the synaptotoxicity induced by the exposure to Aβ1–42.

1–42 (500 nM, 48 h) led to the death of 22 ± 3% of neurons, which displayed apoptotic features (nuclear condensation, cytochrome c release, caspase 3 activation). Blockade of P2Rs with 10 µM PPADS prevented Aβ1–42-induced neurotoxicity, which was mimicked by the blockade of P2YRs with 10 µM reactive blue 2, but not P2X1–3Rs using 10 µM NF023. The P2Y1R antagonist MRS2179 (10 µM) was also neuroprotective. This neurotoxicity was preceded (at 12 h) by a synaptotoxicity and dendritic atrophy (with no neuronal death yet), which was abrogated upon blockade of P2Y1Rs in accordance with the synaptic localization of P2Y1Rs. Interestingly we found an increased density of P2Y1Rs in hippocampal terminals of both rats (Wistar) and mice (C57-BL/6) two weeks after administration of Aβ1–42 (2 nmol, icv) at a time where they displayed a mnemonic deficit (Y maze) and synaptotoxicity (reduced levels of immunoreactivity for synaptic markers such as SNAP-25 and synaptophysin) but no neuronal death (Fluoro-Jade C staining). Finally, KO P2Y1R mice (C57-BL/6) did not display mnemonic deficit or synaptotoxicity two weeks after administration of Aβ1–42.

Altogether, these results show that pharmacological blockade or genetic inactivation of P2Y1 receptors is neuroprotective against synaptotoxicity/neurotoxicity and mnemonic impairment caused by Aβ1–42. This indicates that extracelular ATP may be involved in the development of neurotoxicity caused by Aβ1–42, and prompts considering P2Y1R antagonists as potential candidates to interfere with the early events in AD. (Supported by FCT)

Blood leucocytes are mobile modulators of adenine nucleotide metabolism and play an important role in determining the effects of adenine nucleotides on platelet function

Stan Heptinstall, Susan C Fox, Jacqueline R Glenn, Andrew Johnson, Ann E White, Gerry Dolan1 and Bethan Myers1.

Centre for Integrated Systems Biology and Medicine, University of Nottingham, and 1Department of Haematology, Queen's Medical Centre, Nottingham, UK. s.heptinstall@nottingham.ac.uk

Endothelial cells display the ecto-enzyme NTPDase (CD39) which converts ATP into ADP and ADP into AMP and it is believed that that these vascular cells play an important role in removing adenine nucleotides from blood and preventing ADP causing widespread platelet activation with potential pro-thrombotic consequences. Recently we have made several observations on the importance of leucocytes in adenine nucleotide metabolism and their effects on platelet function. First we found that ATP added to blood induces platelet aggregation although no such aggregation occurs in platelet-rich plasma (PRP). By adding autologous leucocytes back to the PRP we showed that leucocytes are involved in the aggregation response and through the use of selective antagonists showed that ADP is also involved (1). We then found that leucocytes (all neutrophils, all monocytes and a subset of lymphocytes) test positive for CD39. In addition we found modified platelet aggregation responses to both ATP and ADP in association with high leucocyte counts. Aggregation induced by ATP was more rapid while aggregation induced by ADP was followed by rapid disaggregation. This was the case in blood from patients with leucocytosis and also when leucocytosis was created experimentally in normal blood by adding autologous leucocytes (2). A systematic analysis of the role of blood cells and plasma enzymes in adenine nucleotide metabolism by HPLC was then performed. Studies in normal blood confirmed that leucocytes are the principle means through which ATP and ADP added to blood are broken down to ADP and AMP respectively, and that platelets and erythrocytes play virtually no part (3). Studies in leucocytosis demonstrated enhanced rates of conversion of ATP to ADP and of ADP to AMP and explained the effects of leucocytosis on platelet aggregation to ATP and ADP that we had demonstrated earlier (4). Recently we looked further at the role of CD39 on leucocytes in these processes. Blood was obtained from patients with leucocytosis and normal controls, and analyses were performed of platelet aggregation (measured in blood and PRP) and adenine nucleotide metabolism (measured by HPLC). Levels of CD39 on leucocytes were determined by flow cytometry. In all cases abnormalities in platelet aggregation in association with leucocytosis could be explained by enhanced adenine nucleotide metabolism brought about by increased amounts of CD39 on either myeloid or lymphoid cells. In one case in which leucocyte count returned to normal following successful chemotherapy, adenine nucleotide metabolism and CD39 also returned to normal. In summary, leucocytes provide a means of metabolising adenine nucleotides that is additional to that provided by vascular endothelial cells. Their mobility may enable modulation of platelet function throughout the entire vasculature as well as at its periphery.

Boranophosphate Isosteres of Dinucleoside Polyphosphates are Highly Potent and Selective P2Y1-receptor Agonists1

Bilha Fischera, Victoria Nahuma, Mohan Tulapurkarb, Georg Reiserb, Sébastien A. Lévesquec, and Jean Sévignyc bfischer@mail.biu.ac.il

a. Department of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel

b. Institute for Neurobiochemistry, Faculty of Medicine, Otto von Guericke University, Leipziger Str. 44 D-39120 Magdeburg, Germany c. Centre de recherche en Rhumatologie et Immunologie, Université Laval, Sainte-Foy, Québec, Canada.

Dinucleoside polyphosphates, NpnN′, exerting their physiological effects via P2-receptors, are attractive drug targets as they offer better stability and specificity compared to nucleotides. To further improve the properties of NpnN′, which are still pharmacologically unsatisfactory, we developed novel boranophosphate isosters of dinucleoside polyphosphates, denoted as Npn(B)N. These analogues were obtained in a facile and efficient synthesis as the exclusive products in a concerted reaction of two nucleoside phosphorimidazolides and inorganic boranophosphate. This unusual reaction is due to the pre-organization of three reactant molecules by the Mg2+ ion. We found that Ap3/5(βγ-B)A analogues were potent P2Y1-R agonists. Ap5(γ-B)A, was equipotent to 2-MeS-ADP (EC50 6.3 × 10−8 M), thus making it one of the most potent P2Y1-R agonists currently known. Moreover, Ap5(γ-B)A did not activate P2Y2-R. In contrast, Up3/5(βγ-B)U analogues were extremely poor agonists of both P2Y1-R and P2Y2-R. Npn(B)N analogues exhibited remarkable chemical stability under physiological conditions. Under conditions mimicking gastric juice, Np3(γ-B)N analogues exhibited half-life (t1/2) of 1.3 h, whereas Np5(β-B)N degraded at a much faster rate (t1/2 18 min). The hydrolysis of Ap3(β-B)A by human nucleotide pyrophosphatase phosphodiesterases (NPP1 and NPP3) was slowed by 40% and 59%, respectively, as compared to Ap3A. However, this effect of the boranophosphate was position-dependent, as Np5(γ-B)N was degraded at a comparable rate to that of Np5N. In summary, Ap5(γ-B)A appears to be a highly potent and selective P2Y1-R agonist, as compared to the parent compound.

C34T polymorphism in AMPD1 gene potentiates forearm reactive hyperemia in young healthy subjects

NP Riksen1, Barbara Franke2, P Smits1, GA Rongen1

1Dept of Pharmacology-Toxicology, and 2Human Genetics, Radboud University Nijmegen Medical Center, the Netherlands. N.Riksen@aig.umcn.nl


The adenosine mono-phosphate deaminase (AMPD) enzyme converts AMP into IMP in muscle cells. The common variant C34T encodes a severely truncated protein and is associated with improved cardiovascular outcome in patients with coronary artery disease and heart failure, possibly due to increased levels of interstitial adenosine during ischemia. We hypothesize that in subjects with this polymorphism increased adenosine receptor stimulation during ischemia and reperfusion potentiates reactive hyperemia.


In 100 healthy subjects the AMPD1 genotype was determined, as well as nucleoside uptake and adenosine kinase activity in erythrocytes. In 10 individuals heterozygous for the variant allele (C/T) and 10 matched controls (C/C) forearm blood flow (FBF) was assessed with venous occlusion plethysmography during 3, 5 and 5 minutes after 2, 5 and 13 minutes of forearm ischemia, with and without concomitant infusion of the nucleoside uptake inhibitor dipyridamole into the brachial artery (7.4 nmol/min per dl of forearm tissue). Maximum FBF and FBF averaged per consecutive minute were analysed. After washout, vasodilation induced by intrabrachial infusion of sodium nitroprusside (SNP) and acetylcholine (ACh) was determined.


Both groups were similar in age, sex, weight, blood pressure, heart rate, baseline caffeine concentration, and Vmax and Km values for the nucleoside transporter and adenosine kinase. Maximal FBF after 2, 5 and 13 minutes of ischemia was 25.4 ± 2.5, 32.7 ± 2.2, and 38.6 ± 2.6 ml/min per dl of forearm tissue in the C/T group and 21.9 ± 2.2, 28.5 ± 2.4, and 41.0 ± 3.3 ml/min/dl in the C/C group (mean ± SE, P<0.05, repeated measures ANOVA). One-minute averaged FBF was also higher in the C/T group after 2 minutes of ischemia (P<0.05), but not after 5 minutes (P = 0.09) and 13 minutes (P = 0.6). One-minute averaged FBF was potentiated by dipyridamole in both groups for all periods of ischemia (P<0.05). However, the effect of dipyridamole on maximum FBF after 2, 5 and 13 minutes of ischemia was more pronounced in the C/C group than in the C/T group (P<0.05). Vasodilation induced by SNP and ACh did not differ between both groups (P = 0.9 and P = 0.7, respectively).


Dipyridamole potentiates reactive hyperemia, suggesting predominant extracellular formation of adenosine during ischemia and early reperfusion. Heterozygous deficiency in AMPD1 augments reactive hyperemia but reduces the potentiating effect of dipyridamole during maximum vasodilation. These observations support the concept that the cellular uptake of adenosine is reduced in AMPD1 deficient subjects during ischemia and early reperfusion, resulting in increased adenosine receptor stimulation. This mechanism could well contribute to the observed improved cardiovascular outcome in patients with the C34T polymorphism of the AMPD1 gene.

Ca2+-dependent release of adenine and uridine nucleotides from A549 cells

Sabina Tatur1, Silvia Kreda2, Eduardo Lazarowski2 and Ryszard Grygorczyk1

1 Research Centre, Centre hospitalier de l'Université de Montréal (CHUM) — Hôtel-Dieu, and Department of Medicine, Université de Montréal, Montréal, Québec, Canada

2 Cystic Fibrosis Center, University of North Carolina School of Medicine, Chapel Hill, NC 27599-7248, U.S.A. ryszard.grygorczyk@umontreal.ca

Extracellular nucleotides play an important role in airway defense by controlling epithelial ion and fluid transport as well as ciliary beating. Nucleotide levels on airway surfaces, measured in vitro and ex vivo, show dynamic changes due to the combination of basal and stimulated release and their rapid metabolism. The mechanism of nucleotide release from epithelial cells is not well understood, and their hydrolysis at the airway surface makes it difficult to assess the magnitude and relative abundance of different nucleotide species released. In this study, to minimize cell surface hydrolysis, we used a low-volume (300-µL) flow-through chamber (1.3 ml/min perfusion rate), and examined adenine and uridine nucleotide content in perfusates of human A549 cells prior to and during 50% hypotonic shock. Aliquots of the perfusates were collected at 15- to 60-s intervals. ATP, ADP, AMP, and Ado were quantified by HPLC analysis of fluorescent etheno derivatives, and UTP and UDP were measured using HPLC-coupled radioenzymatic assays. After the onset of hypotonic shock, ATP, ADP, UTP, and UDP content in the perfusates increased markedly and peaked at approximately 2 min, followed by a gradual decay during the next 15–20 min; changes in Ado and AMP at peak were relatively minor. The peak concentrations and fold increment (in brackets) were: 52 nM ATP [5], 16 nM ADP [5], 4 nM AMP [2], 27 nM Ado [1.3], 21 nM UTP [>7], and 11 nM UDP [12]. Nucleotide release was almost completely abolished from cells loaded with the calcium chelator BAPTA. Under isotonic conditions, elevation of intracellular calcium with the calcium ionophore ionomycin (5 µM, 3 min) also released nucleotides with kinetics and relative abundance as above, although less robustly. ATP:ADP (3:1) and UTP:UDP (2:1) ratios in the perfusates from stimulated cells were markedly smaller than the cytosolic ratios of these species, suggesting that an NDP-rich compartment, e.g., the secretory pathway, contributed to nucleotide release. Laser confocal microscopy experiments illustrated increased uptake of FM1-43 into the plasma membrane upon hypotonic shock or ionomycin treatment, consistent with enhanced vesicle exocytosis under these conditions. In summary, our results strongly indicate that calcium-dependent exocytosis is responsible, at least in part, for adenine and uridine nucleotide release from A549 cells.

Supported by the Canadian Institutes of Health Research (to R.G.) and the National Institutes of Health USA (to E.L.).

Caffeine intake induces an alteration of human neutrophil A2A adenosine receptors

Katia Varani1, Francesco Portaluppi1, Stefania Gessi1, Stefania Merighi1, Fabrizio Vincenzi1, Annalisa Benini1, Elena Cattabriga1, Alessandro Dalpiaz1, Fabrizio Bortolotti1, Luiz Belardinelli2, Pier Andrea Borea1

1Department of Clinical and Experimental Medicine and the Department of Pharmaceutical Sciences, University of Ferrara, Italy and the 2Department of Medicine, CV Therapeutics Inc, Palo Alto, CA.

Caffeine is the most widely used drug in the world and acts mainly through antagonism of the effects mediated by the adenosine receptor subtypes A1, A2A, A2B and A3. Controversial results have been reported on the link between caffeine and inflammation. Recently, caffeine has been found to have anti-inflammatory activity in different human substrates. From this background the present study was designed to evaluate the effect of repeated caffeine administration at different doses and for different periods of time (400 or 600 mg/d for 1 week and 400 mg/d for 2 weeks) in human neutrophils. Blood and plasma levels of methylxanthines revealed no significant concentrations after caffeine abstinence. Neutrophils were obtained from peripheral venous blood of 33 healthy human volunteers at the end of 2 weeks of caffeine abstinence and at 1, 12, 24, 36, 48, 60 hours after the last dose of caffeine. Saturation binding assays showed an increase of affinity (KD) and density (Bmax) of A2A adenosine receptors after caffeine intake at the doses of 400 mg/d for 14 days or 600 mg/d for 7 days. Interestingly, the increase in density of A2A receptors found at 1 hour after caffeine withdrawal was similar to that observed at 12, 24 and 36 hour after the last dose of caffeine. On the other hand at 48 and 60 hour after the last dose of caffeine the binding parameters were not different from control condition or from the control group. The upregulation of A2A receptors does not appear to be ascribable to the synthesis of new receptors because were no changes in mRNA levels at different times after the last dose of caffeine. The A2A receptor alteration was accompanied by increases in cAMP accumulation and decreases in superoxide anion production after A2A stimulation by NECA. Binding and functional changes of A2A receptors gradually return to baseline after 48 hours of caffeine withdrawal. In summary, caffeine intake and subsequent withdrawal mediates a temporary upregulation of A2A receptors and as a consequence might represent the mechanism responsible for the observed anti-inflammatory effects.

Caffeine modulates P50 auditory sensory gating in healthy subjects

Diogo R. Lara1, Eduardo S. Ghisolfi1, 2, Alice Schuch2, Ivo M. Strimitzer Jr.2, Gustavo Luersen1, Fabíola F. Martins1, Fernanda L. P. Ramos1, Jefferson Becker1

1 Departamento de Ciências Fisiológicas, Faculdade de Biociências, PUCRS, Porto Alegre, Brazil; 2Departamento de Bioquímica, ICBS, UFRGS, Porto Alegre, Brazil. drlara@pucrs.br

The P50 suppression paradigm is an index of sensory gating assumed to reflect an inhibitory process. Adenosine is a neuromodulator with mostly inhibitory activity that is released by physiological stimuli and can be blocked by non-selective adenosine receptor antagonists such as theophylline and caffeine. A previous study showed that a single dose of theophylline decreased P50 suppression in healthy volunteers. Here we investigated the effect of caffeine (0, 100, 200 and 400 mg p.o.) on P50 sensory gating in 24 healthy volunteers (15 habitual caffeine highusers and 9 low-users). The 200 mg and 400 mg doses reduced P50 sensory gating (increase P50 ratio), whereas 100 mg produced a non-significant effect. The effect of caffeine on P50 ration was independent of gender and habitual caffeine intake. High caffeine users also showed baseline differences, with lower S2 amplitudes compared to low-users. These results reinforce the participation of adenosine in the modulation of P50 sensory gating and suggest that both chronic and acute caffeine ingestion should be controlled for in studies using the P50 sensory gating paradigm.

Caffeine recognizes differently adenosine A2A receptors in homo or heterodimers: A mechanism for caffeine tolerance

Gemma Navarro, Francisco Ciruela, Sergi Ferré, Carmen Lluis, Rafael Franco Vicent Casadó.

Molecular Neurobiology Unit. Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS). University of Barcelona. Diagonal 645. 08028 Barcelona. rfranco@ub.edu

Caffeine is the most world-wide consumed psychoactive drug and its only known molecular targets at non-pathological doses are adenosine A1 and A2A receptors. Of the four known adenosine receptors (A1, A2A, A2B and A3), adenosine A1 (A1Rs) and A2A (A2ARs) receptors are primarily responsible for the central effects of adenosine. The molecular mechanisms underlying the effects of caffeine are not fully elucidated. As indicated in another session of this Conference (Franco R) and in Ciruela et al., (2006), A1 adenosine receptors form heteromers with adenosine A2A receptors. Functional studies showed that A1R-A2AR heteromers are responsible for a strong A1R-A2AR antagonistic cross-talk. The strength of this cross-talk would depend on the concentration of extracellular adenosine. To illustrate this, it can be expected that at adenosine concentrations high enough to activate A2AR, the signalling via A1R is severely impaired. In experiments of displacement of radioligand binding in transfected cells or in samples from striatum, it was found that A1R showed the same KD for caffeine in A1RA2AR co-transfected than in A1R-transfected cells. Also, as previously shown using other transfected cell lines, A2ARs displayed higher affinity (lower KD value) for caffeine than A1Rs in single-transfected HEK cells. In contrast, the affinity of caffeine for A2ARs in A1R-A2AR co-transfected cells was 12-times higher (p<0.001) than in A2AR-transfected cells. The affinity of the A2AR for caffeine was not altered when co-transfected with the dopamine D2 receptor (D2R), even though they form A2AR-D2R heterodimers. These results demonstrate the existence of a selective reduction in the affinity of A2AR for caffeine in the A1R-A2AR heteromer, i.e. the binding of caffeine to A2ARs in the A1R-A2AR heteromer is qualitatively different from the binding to the non-heteromeric A2AR or other heteromeric A2ARs.

An unresolved issue about caffeine is the strong tolerance for many of its behavioural and biochemical effects that develops after chronic treatment. It is generally assumed that tolerance to the behavioral effects of caffeine is mostly related to pharmacodynamic factors involving adenosine receptors, but there is no consensus about which are the significant changes in either A1R or A2AR function after chronic caffeine treatment. The most commonly reported effect is the up-regulation of A1Rs. Nevertheless, there are reports indicating a lack of significance in these changes in receptor density and there are even studies showing down-regulation of either A1R or A2AR function. To test the role of A1R-A2AR heteromers in this phenomenon, competitive-inhibition experiments of [3H]R-PIA binding using CGS21680 were performed in striatal membrane preparations of caffeine-treated rats. The displacement curve of [3H]R-PIA binding by CGS21680 in striatal membranes from caffeine-treated rats was significantly (p<0.001) better fitted by a two-site model than by a single-site model. The IC50 value corresponding to the binding of CGS21680 to A1R (1.7 ± 0.5 µM) was similar to that obtained in naïve rats (1.8 ± 0.5 µM). However, the inhibition of the [3H]R-PIA binding to A1R when CGS21680 binds to the A2AR displays an EC50 value (8 ± 3 nM) about three fold lower than the value obtained in naïve rats (22 ± 6 nM). Interestingly, these results indicate that caffeine pre-treatment alters the function of the A1R-A2AR heteromers, increasing significantly (p<0.02) the sensitivity of A2AR to modulate (negatively) A1Rs.

Apart from the affinity of the A1R, A2AR and A1R-A2AR for adenosine and caffeine and from the strength of the A1R-A2AR intramembrane interaction, one more variable plays a substantial role when analyzing the effects taking place after chronic caffeine administration. In fact chronic treatment with the methylxanthine leads to a significant increase in the plasma and extracellular levels of adenosine. Since at relatively low adenosine levels there is little occupancy of A2ARs most of the behavioral and biochemical effects following an acute administration of caffeine are due to A1R blockade. Under chronic caffeine treatment (or other conditions leading to increased adenosine levels), adenosine also binds and activate A2AR, which, in addition, has a reduced affinity for caffeine. This likely scenario would lead to a situation where the increased levels of adenosine acting on A2ARs to inhibit A1R function by means of the potentiated A1R-A2AR intramembrane interaction. Under these conditions, caffeine would have little effect on A1Rs, which would be already inhibited as a consequence of the A1R-A2AR intra-membrane inter-molecular interaction.

Can Acadesine increase adenosine level in rat cardiomyocytes?

Jerzy Barankiewicz and Piotr Chomczynski

Molecular Research Center, Inc, Cincinnati, OH 45212 jbarankiewcz@mrcgene.com

Acadesine (AICA-riboside) is thought to have a protective effect during cardiac ischemia via two mechanisms. Acadesine can protect energy metabolism by elevating adenine nucleotides concentration, or it can elevates adenosine (Ado) concentration, which acts via Ado receptors. In this study, we evaluated both of these mechanisms in isolated, spontaneously beating rat cardiomyocytes and neonatal rat heart slices under normoxic and various stress conditions.

In isolated rat cardiomyocytes and heart slices under normoxic conditions, 500 µM radiolabeled acadesine was efficiently metabolized to the purine nucleotides ATP and GTP, although Hyp was the predominant product. However, as acadesine rose to surpass 1 mM concnetration, radiolabeled ZMP accumulation increased and exceeded the ATP concentration.

After 2 hrs incubation under anaerobic conditions in an argon atmosphere, ATP levels decreased three-fold in isolated rat cardiomyocytes. In the same cells transferred to aerobic conditions in the presence of 500 µM acadesine, the ATP level appropriate for normoxic conditions was reconstituted after 2–3 hours incubation. Further incubation in this condition caused a continuing increase in ATP levels.

While no Ado accumulation was observed under the conditions described above, we also evaluated Ado accumulation in the presence of acadesine while chemically accelerating the cascade of ATP degradation. When antimycin-A or iodoacetate chemically stressed cardiomyocytes or heart slices, Hyp and inosine increased several-fold but not Ado accumulation was observed. However in the presence of 2′-deoxycoformycin — inhibitor of ADA activity, Ado accumulated but acadesine only slightly increased Ado concentration.

As opposed to Ado release and accumulation, we examined the effect of acadesine on Ado uptake in rat cardiomyocytes. Acadesine inhibited Ado uptake in these cells in a dose-dependent fashion under both normoxic and anaerobic conditions. This effect required however a relatively high concentration of acadesine.

In conclusion, these data indicate that acadesine treatment effectively supports reconstitution of ATP levels ischemic stress but it has no effect on Ado release and accumulation in rat cardiomyocytes. However, acadesine may affect Ado uptake in the same cells.

Cardiac expression of NTPDase1 and caveolins are altered in human disease

Andrá s Bodnár1, Péter Pócza1, Ida Matkó1, Beáta Sperlágh2, Anna L. Kiss1, Ágnes Kittel2

1Medical School, Semmelweis University, Budapest, Hungary, 2Institute of Experimental Medicine, Hungarian Academy of Sciences, Budapest, Hungary kittel@koki.hu

Pathological circumstances like inflammation or ischemic insult facilitate the release of adenine nucleotides among ATP, from several types of cells. Concentration of extracellular ATP is regulated by ectonucleotidases. Endothelial and smooth muscle cells possess high ecto-ATPase activity and it has been demonstrated that this enzyme activity is altered after LPS stimulus, ischemia, inflammation, viral infection and in multidrug resistance. Upregulation of increased ecto-ATPase activity is associated with the appearance of higher number of caveolae, specialized membrane invaginations, not only in endothelial and smooth muscle cells but, also in pericytes, astrocytes and multidrug resistant cancer cells. Several molecules involved in signalling (e.g. eNOS, G-proteins, receptors) are targeted to these membrane compartments; thus caveolae appear to integrate cellular activation events. The ecto-ATPase enzyme within these signal-transducing microdomains has been identified as NTPDase1/CD391. NTPDase1/CD39 can co-associate with scaffolding domain of caveolin1 or 3, or of RanBPM, and other structural proteins of caveolae. The question has been arisen whether there is a link between higher NTPDase expression and expression of caveolins during disease. We demonstrated the ecto-ATPase activity of healthy and diseased human cardiac tissue by enzyme histochemistry, measured the concentration of the extracellular ATP and metabolites by HPLC and identified NTPDase1/CD39 by immunohistochemistry and Western blotting2. The presence of all the three types of caveolins were demonstrated in parallel ultracryo sections. We demonstrated the presence of caveolin1 in human cardiac muscle cells. Double immunostaining confirmed the co-localization of NTPDase1/CD39 and caveolin1 in endothelial cells, whereas the CD39 signal was weak in the cell membrane of cardiac muscle cells. Pathological samples from patients with ischemic heart disease exhibited significant increases in ecto-ATPase activity, as measured by HPLC and RT-PCR. Western blotting confirmed higher CD39 expression in diseased as compared with control tissues. This higher activity and expression level of NTPDase1 underline a putative protective roles in the cardiovascular system and also support the administration of soluble NTPDase1/CD39 in the treatment of ischemic complications. Caveolin3 expression was altered in pathological samples. The lower expression level of caveolin3 may indicate lability in contrast to caveolin1.The high level presence of caveolin1 in diseased cardiac muscle raises the possibility, that this caveolin isoform, at least in part, may substitute for caveolin3.

Research was supported by Hungarian grants ETT 480/2003, OTKA M036314.

CD73-deficient mice show increased lymphocyte migration through high endothelial venules (HEV) in draining lymph nodes during inflammation

Masahide Takedachi1, Yukihiko Ebisuno2, Stephanie McGee1 and Linda F. Thompson1

1Immunobiology and Cancer Program, 2 Cardiovascular Biology Program, Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma, USA. Masahide-Takedachi@omrf.ouhsc.edu

CD73 is a GPI-anchored cell surface protein with ecto-5′-nucleotidase enzyme activity which catalyzes the dephosphorylation of 5′-adenosine monophosphate to adenosine. As this molecule catalyzes the last step in the generation of adenosine from extracellular adenine nucleotides, it has the capacity to regulate the activation of adenosine receptors. For example, adenosine generated by CD73 can decrease endothelial cell permeability in in vitro1). Moreover, Thompson et al.2) demonstrated that cd73-deficient mice have defects in endothelial barrier function leading to neutrophil accumulation in tissues during hypoxia, a condition in which CD73 expression is up regulated. In this study, we asked whether CD73 might also have a role in regulating lymphocyte migration into lymph nodes during inflammation. Lymphocytes home from the blood stream to lymph nodes (LN) through a process of tethering, rolling, adhesion and transmigration through the vasculature. High endothelial venules (HEV) are a key regulator of this process. The interaction of L-selectin on lymphocytes with peripheral-node addressins on HEV initiates tethering and rolling. Then, activation of β2 integrin (LFA-1) on lymphocytes by chemokines induces firm adhesion with HEV. However little is known about the transmigration process and the factors that control HEV permeability. Here we show preliminary evidence for a role of CD73 in regulating the entrance of lymphocytes into draining LN under inflammatory conditions induced by LPS and poly (I:C). The high expression of CD73 on HEV has been already reported3). The size of secondary lymphoid organs is normal in cd73-deficient mice, making it likely that lymphocyte homing to peripheral LN is also normal under steady state conditions. To examine the role of CD73 in lymphocyte migration during an inflammatory response, cd73-deficient mice and wild type controls were subjected to subcutaneous injection of LPS or poly(I:C) into the left hind footpad, and lymphocyte migration into the draining popliteal lymph nodes was examined. PBS was injected into contralateral site as a control. As expected, 24 hours after injection, the left draining lymph nodes were dramatically enlarged in both strains of mice (p<0.02); however, they were approximately twice as large in the cd73-deficient mice as in the controls (p<0.04). To more directly assess lymphocyte migration, fluorescently labeled wild type splenocytes were injected i.v. 24 hours after LPS or poly (I:C) injection and the accumulation of labeled lymphocytes in the popliteal lymph nodes was measured after 1 hour. Although no difference was observed on the control side, the number of lymphocytes that migrated into the draining LN of cd73-deficient mice was 2.5-fold increased compared with that of wild type mice (p<0.01). These results suggest that CD73 on HEV modulates lymphocyte migration into peripheral draining lymph nodes during inflammation. Experiments are in progress to identify the specific adenosine receptors involved.

Cells and drugs and rac and rho: A mechanistic look at chemotaxis mediated by the P2Y2 receptor

Laurie Erb, Zhongji Liao, Sriparna Bagchi, Cheikh I. Seye & Gary A. Weisman

Department of Biochemistry, University of Missouri, Columbia, MO 65211 USA erbl@missouri.edu

Chemotaxis, or directed cell migration, is a fundamental feature of eukaryotic cells and is important for many physiological events, including embryonic development, inflammation and wound healing. The ability of a cell to undergo chemotaxis requires the cell to assume a polarized morphology that is controlled by cell surface receptors that activate the Rho family of small GTPases, including Cdc42, Rac and Rho [1,2]. Upon activation of a chemoattractant receptor, Cdc42 and Rac localize at the leading edge of a cell and control directional cell movement and lamellipodium formation, respectively [1]. Rho localizes at the rear and sides of a cell and controls the formation of contractile actin-myosin stress fibers [3]. Together, these GTPases promote cell migration towards a chemoattractant by mediating extension of the actin cytoskeleton at the front edge of the cell and retraction of the cytoskeleton at the rear. Recent studies have shown that G protein-coupled receptors activate Rac and Racdependent lamellipodia formation through Gi/o, whereas the activity of Rho and Rho-dependent stress fiber formation is controlled by G12/13 [3]. Previously, we reported that the G protein-coupled P2Y2 receptor (P2Y2R) contains a consensus Arg-Gly-Asp (RGD) integrin-binding sequence that interacts with ανβ3 and ανβ5 integrins [4]. Furthermore, we found that the RGD domain in the P2Y2R is required for Go- but not Gq-mediated calcium signaling [4], leading us to speculate that αν integrin interaction with the P2Y2R is important for nucleotide-induced chemotaxis. In this study, we show that mutation of the RGD sequence to RGE in the human P2Y2R expressed in 1321N1 astrocytoma cells completely prevents UTP-induced chemotaxis and stress fiber formation as well as activation of Go, G12, Rac, Rho and Vav2, a guanine nucleotide exchange factor for both Rac and Rho. In cells expressing the wild-type but not the RGE mutant P2Y2R, UTP also increases expression of vitronectin, an extracellular matrix protein that is a ligand for ανβ3/5 integrins. P2Y2R-mediated chemotaxis, Rac and Vav2 activation, and vitronectin upregulation were inhibited by pretreatment of the cells with anti-ανβ5 integrin antibodies, αν antisense oligonucleotides or the Gi/o inhibitor, pertussis toxin. P2Y2R-mediated stress fiber formation and Rho activation were inhibited by anti-ανβ5 integrin antibodies, αν antisense oligonucleotides or a dominant negative Gα12 construct. Collectively, these results suggest that the P2Y2R requires interaction with αν integrins for coupling to Go and G12 proteins involved in chemotaxis.

This study was supported by the National Institutes of Health Grants 1 P01-AG-018357 and 1 R01-DE-07389.

Cerebral Arteriolar Response to Neuronal Activation is Attenuated Following Subarachnoid Hemorrhage

Joseph R. Meno1, Ik-Seong Park1, Taylor J. Abel1, Abhineet Chowdhary1, Thien-son K. Nguyen1, H. Richard Winn2, Al C. Ngai1 & Gavin W. Britz1

1Department of Neurological Surgery, University of Washington, Seattle, Washington 98104

2Department of Neurosurgery, Mount Sinai School of Medicine, New York, New York 10029 richard.winn@mountsinai.org

Small diameter arterioles play a significant role in the maintenance of cerebral blood flow (CBF). However, the status of cerebral arteriolar reactivity and neurovascular coupling following SAH is unclear. The present study tests the hypothesis that SAH results in alterations in cerebral arteriolar response to neuronal activation. An endovascular filament model was used to induce SAH in halothane-anesthetized male Sprague-Dawley rats. Then, at 24, 48, 72, 96, and 120 hours post-SAH, pial arteriolar responses to contralateral sciatic nerve stimulation (SNS) were evaluated utilizing a closed cranial window technique. In addition, somatosensory evoked potentials (SEPs), CO2 reactivity, as well as dose-responses to topical application of adenosine (ADO) and sodium nitroprusside (SNP) were assessed. In sham-operated rats, SNS evoked a 23.6 ± 1.8% increase in arteriolar diameter. In rats subjected to SAH, pial arteriolar response to SNS was significantly attenuated to 13.7 ± 0.9%, 11.9 ± 1.3% and 15.2 ± 1.2% at 24, 48 and 72 hours post-SAH, respectively (p < 0.05; n ≥ 7). By 96 and 120 hours post-SAH, SNS-induced dilations recovered and were similar to sham responses. In contrast, SEPs were unaffected, suggesting a direct effect of SAH on cerebral arterioles that cannot be attributed to alterations in evoked neuronal or metabolic activity. The effect of SAH on arteriolar reactivity to ADO or SNP paralleled our observations during SNS. Pial vasodilatation to ADO (10 µM) and SNP (1 µM) were significantly attenuated by 47% and 41%, respectively, at 48 hours post-SAH (p < 0.05; n = 7), whereas responses at 96- and 120 hours post-SAH were similar to sham. SAH also had no effect on CO2 reactivity suggesting that the attenuation of SNS-induced vasodilatation following SAH is not the result of non-specific cerebral arteriolar paralysis. The present study demonstrates that SAH significantly attenuates cerebral arteriolar response to neuronal activation; the effects of SAH peak at 48 hours and return to normal by 96 hours post insult. Moreover, vasodilatation to ADO and SNP (NO), purported mediators of CBF regulation during increased neuronal activity, were similarly attenuated by SAH. The results of the present study provide the first evidence for alterations in arteriolar reactivity during increased neuronal activity and underscore the importance of future investigations aimed at evaluating the status of cerebral arteriolar reactivity following SAH.

Acknowledgements: This study was supported by grants from the Research Foundation of the American Association of Neurological Surgeons (GWB), the American Heart Association (ACN) and the NIH (HRW, NS-21076)

Challenges and Oppurtunities in the Identification of Insulin-Like Growth Factor-I Receptor Kinase Inhibitors

Carlos Garcia-Echeverria

Novartis Institutes for BioMedical Research — Oncology Research, CH-4002 Basel, Switzerland carlos.garcia-echeverria@novartis.com

The Insulin-like Growth Factor-I Receptor (IGF-IR) is a member of the insulin receptor family of tyrosine kinases. A broad range of experimental studies have revealed that IGF-IR function is implicated in most of the hallmarks of cancer, but it is probably the anti-apoptotic activity of this receptor that makes its kinase activity an attractive therapeutic target in anti-cancer drug discovery. In this context, the identification of specific low-molecular mass inhibitors of IGF-IR has proven to be a major challenge for medicinal chemistry due to the high sequence identity at the kinase domains of IGF-IR and InsR (around 84%) and, in particular, at the ATP-binding pocket (100% sequence identity). This presentation will cover the identification and characterization of a new series of IGF-IR kinase inhibitors. The selectivity achieved at the cellular level with these compounds suggest conformational differences between the native forms of IGF-IR and InsR from the unactivated to the fully activated form, that can effectively be exploited for drug discovery.

Changes in levels of 3′-AMP, an intracellular P-site inhibitor of adenylate cyclase, and its forming enzyme activity in diabetic mice treated with insulinomimetic zinc (II) complex

Akihiro Miyamoto1, Hiroyuki Fujimori1, Takahiro Horiuchi1, Hiromu Sakurai2 & Hidemitsu Pan-Hou1

1Department of Analytical Chemistry in Hygiene, Faculty of Pharmaceutical Sciences, Setsunan University;

2Department of Analytical and Bioinorganic Chemistry, Kyoto Pharmaceutical University 05d101ma@edu.setsunan.ac.jp

Adenylate cyclase(AC) is regulated by a number of extracellular and intracellular signals. Intracellular adenosine 3′-monophosphate (3′-AMP), one of the degradation products of RNAs, is pharmacologically classified as a P-site inhibitor of AC. The 3′-AMP forming enzyme, one of the RNase, has been shown to exist in various organs of rat and mice [1,2]. The forming enzyme activity is shown to be inhibited by divalent metal ions, especially zinc ion, and activated by EDTA. However, the biological roles of 3′-AMP and its forming enzyme still remains unclear. Recently, Fujimori and Pan-Hou [3] reported that 2,5-dideoxyadenosine, a P-site inhibitor of AC, enhanced cellular ATP levels in PC 12 cells. As cAMP released by AC modulates glycolysis and insulin secretion, 3′-AMP and its forming enzyme might be involved in glucose metabolism in various organs including pancreas and liver. Streptozotocin (STZ) induces experimental diabetes in animals. Bis(picolinate)zinc (II) [Zn(pa)2] has been known to possess potent insulinomimetic action of lowering blood glucose levels [4]. Zinc ion is not only an essential trace element but also has inhibitory effect on 3′-AMP forming enzyme activity. Therefore, it is of interest to know whether 3′-AMP levels and its forming enzyme activity in STZ-induced diabetic mice might be affected by the treatment of Zn(pa)2. The objective of our reserch was to examine the effect of Zn(pa)2 on 3′-AMP levels and its forming enzyme activity in STZ-induced diabetic mice.

Male ICR mouse were treated with intraperitoneal injections of STZ (40 mg/kg body weight) during five consecutive days, and then Zn(pa)2 (10 mg/kg body weight × 2 days and 5 mg/kg body weight × 9 days) was intraperitoneally injected [4]. Determination of 3′-AMP and its forming enzyme activity by HPLC were performed according to the method described by Fujimori and Pan-Hou [1].

Zn(pa)2 effectively lowered the STZ-induced enhancement of blood glucose levels. Pancreatic 3′-AMP levels in the STZ-, Zn(pa)2- and STZ- Zn(pa)2 groups decreased, and hepatic levels in the Zn(pa)2 and STZ- Zn(pa)2 groups increased. Pancreatic 3′-AMP forming enzyme activities in the Zn(pa)2 and STZ- Zn(pa)2 groups and hepatic the activities in the Zn(pa)2 group increased. These results suggested that 3′-AMP metabolism in pancreas and liver in the STZ-induced diabetes at early stage might be affected by the treatment of Zn(pa)2.

Channel activity using the consensus segment of the M2 transmembrane domain from P2X7 receptor

Cristina Alves Magalhães de Souza1,1,2, Pedro Teixeira3, Robson Xavier Faria1, Oxana Krylova2, Peter Pohl2 & Luiz Anastacio Alves1

1Oswaldo Cruz Foundation — Dept. Immunology, Institute Oswaldo Cruz-Av. Brazil 4365-21045-900 RJ, Brazil.

2Research Institute for Molecular Pharmacology — Robert Rö ssler-Strasse 10, 13125 Berlin, Germany. 3 Research Center — Monkida Foundation — Grajau, RJ, Brazil souzacam@ioc.fiocruz.br


P2X7 receptor modulates a spectrum of cellular events in a variety of cells of the immune system. Although the pharmacology and channel properties of the P2X7 receptors have been studied intensively, the formation of the pore associated with the P2X7 remains an open question. Therefore, in the present study, we have investigated the eletrophysiological characteristics of the M2 transmembrane domain portion from the P2X7 receptor.

Methods and Results

Most of bioinformatics studies have shown that M2 transmembrane domain (TM2) from the P2X7 receptor is conformed as an alpha helix. Using new algorithms of bioinformatics and molecular dynamics, we verified that the consensus segment of TM2 might be conformed as a beta sheet configuration. To test this hypothesis, we verified experimentally in subsequent analyses whether this segment would be able to form a channel in a planar lipid bilayer. The segment was synthetized with high degree of purity and its activity onto the planar lipid bilayer was measured using voltage-clamp condition. We observed single-channel currents which had conductance of 10pS in symmetric conditions with KCl, being cation selectivity. This peptide forms also channels in a mammalian cell line (HEK) when analyzed by patch clamp in cell attach configuration.


The consensus peptide segment from the TM2 forms a channel that has characteristics similar to P2X7 channel but different from P2X7 associated pore. In addition, this study raises the possibility that some regions of TM2 are not in alpha helix configuration.

Financial Support


Characterisation of functionally important residues in mouse P2X7 receptor

Pablo Pelegrin, Mark Young & Annmarie Surprenant

University of Sheffield, Department of Biomedical Science, Florey Building, Western Bank, Sheffield, UK, S10 2TN p.pelegrin@sheffield.ac.uk

The ATP-gated P2X7 receptor is an unusual ion channel that couples to multiple downstream signalling cascades. Mouse P2X7, cloned from NTW8 microglial cells and heterologously expressed in HEK293 cells, is much less sensitive to ATP than the human or rat homologs, although in vivo experiments indicate that the endogenous mouse receptor is highly sensitive to ATP. We found single nucleotide variations in mouse cDNA sequences cloned from NTW8 microglial cells [T33,G661,T848] and C57BL/6 mice [G33,A661,C848]. These nucleotide variations result in three amino acid differences; [Phe11,Ala221,Met283] from NTW8 microglia and [Leu11,Thr221,Thr 283] from C57BL/6 mice. We expressed these receptors in HEK293 cells and compared membrane currents, ethidium uptake and surface membrane expression. We found that the presence of Met283 in the sequence derived from NTW8 cells caused massive impairment of the ATP response. Maximum current densities at mouse P2X7 Met283 were <5% of those at mouse P2X7 Thr283, without change in the agonist concentration-response curve or receptor trafficking to the plasma membrane. Moreover, no ethidium uptake was observed for mouse P2X7 Met283. The corresponding mutation in rat P2X7 (Thr283-Met) also yielded currents that were <5% of wildtype and no ethidium uptake was observed. Furthermore, characterization of the mouse P2X7 cytoplasmic domain polymorphism Pro451/Leu451 surprisingly revealed similar maximum current densities and agonist EC50 values for both alleles. However, both ethidium uptake and agonist-induced rise in intracellular calcium concentration were significantly reduced at the [Thr283,Leu451]P2X7 receptor. These results show that Thr283 in the ectodomain is critical for P2X7 receptor function and suggest that the intracellular residue at position 451 may affect downstream signalling independently of ion channel activity.

Characterisation of P2X receptor expression and inflammatory function in human endothelial cells

Heather L. Wilson1, Richard W. Varcoe2, Leanne Stokes2, Sheila E. Francis1, Steven K. Dower1, David C. Crossman1 & Annmarie Surprenant2

1School of Medicine and Biomedical Science, 2Department of Biomedical Science, University of Sheffield, S10 2JF. U.K. H.L.Wilson@sheffield.ac.uk

Interleukin-1 (IL-1) is the prototypical pro-inflammatory cytokine, functioning at the apex of a cascade effecting and co-ordinating a wide variety of inflammatory actions. Evidence from pathological specimens, animal models, genetic and clinical studies implicates IL-1β, particularly of endothelial cell origin, in the pathogenesis of the atherosclerotic plaque and the arterial healing response to injury. The net effect of IL-1 signalling depends primarily on the balance between IL-1β and its naturally occurring receptor antagonist (IL-1ra) acting on type I IL-1 receptors (IL-1R1) at the target cell surface. IL-1β is a so-called leaderless secretory protein in that it lacks a signal peptide sequence directing it to the endoplasmic reticulum and Golgi apparatus for classical exocytotic release. In monocytes, release of leaderless IL-1β requires activation of the P2X7 receptor [1]. More recently, release of the leaderless intracellular isoform of IL-1ra (icIL-1ra), the only isoform produced in endothelial cells, was shown to occur via a P2X dependent process [2]. We therefore determined whether a similar mechanism exists for the release of IL-1β from human endothelial cells.

Expression studies using RT-PCR and qRT-PCR showed that human umbilical vein endothelial cells (HUVECs) predominantly express the P2X4 and P2X7 purinergic receptor subtypes and that their expression is inducible under inflammatory conditions with IFN+ containing combinations having the most pronounced effect. Treatment for 48 hr with IFNγ and TNFα resulted in a 100-fold and 65-fold increase in P2X4 and P2X7 mRNA levels respectively, relative to an untreated control. These changes in mRNA expression were confirmed at the protein level using western blotting, immunoprecipitation and FACS.

Low level functional P2X4 and P2X7 responses were measured in HUVECs by whole cell patch-clamp recording with an increase in the mean current density recorded from IFNγ/TNFα treated cells (1.05 ± 0.28 pA in control and 2.63 ± 0.39 pA in treated HUVECs).

Intracellular pro-IL-1β synthesis is negligible in untreated HUVECs, whereas significant amounts are produced in response to inflammatory stimulation (up to 4000 pg/ml following treatment with IFNγ/TNFα, LPS and BzATP). In contrast, caspase-1 is constitutively expressed in HUVECs but is similarly induced in response to inflammatory stimulation. Low-level release of biologically active 17 KDa IL-1β was measured upon BzATP stimulation of HUVECs by ELISA and Western blots and was in the absence of cell death as shown by an LDH release assay. However, icIL-1ra was released under the same conditions, and at higher concentrations, so the net biological effect is generally anti-inflammatory, when measured using an IL-1 sensitive bioassay.

We conclude therefore, that P2X receptors play an important role in regulating secretion of leaderless pro- and anti-inflammatory cytokines in the endothelium; their expression and activation may determine the balance of IL-1β and icIL-1ra released from endothelial cells, providing a target for altering the inflammatory state of the arterial vessel wall.

Characterisation of a recombinant soluble apyrase from the blood-feeding sand fly Phlebotomus papatasi and anti-thrombotic activity in a rat model of acute thromboembolism

Thomas M. Smith1, Qi Zheng2 & Min Lu2

1Department of Cardiovascular and Metabolic Diseases, Wyeth Research, Cambridge, MA, USA

2Department of Biochemistry, Weill Medical College of Cornell University, New York, New York, USA

Adenosine nucleotides such as ADP found in the blood as a consequence of vascular injury and granule release from activated platelets are one of the most important physiological agonists of platelet recruitment, aggregation and thrombus formation. As such, the blockade of platelet P2Y12 ADP receptors by thienopyridine derivatives (e.g. clopidogrel) and the metabolic removal of ADP from the vascular milieu by ADP-hydrolyzing enzymes (apyrases) offer great advantages for the prevention and treatment of cardio- and cerebrovascular vascular events characterized by activated platelets.

We have previously reported the purification and crystal structure of a novel soluble human apyrase (see Dai et al., 2005. Cell. 116(5): 649–59). Based upon sequence homology to the apyrases identified in blood-feeding arthropods, the human enzyme was “re-engineered” using site-directed mutagenesis to alter the calcium binding and nucleotide substrate active site. As a result, the ADP hydrolyzing activity was enhanced more than 100 fold and this engineered human apyrase was shown to be an effective inhibitor of ADP and collagen induced platelet aggregation.

In spite of the enhanced ADPase activity of the engineered human apyrase however, the protein had lower Kcat and anti-platelet activity compared to the soluble salivary apyrases of blood-feeding arthropods. As a result, we expressed and purified the recombinant soluble apyrase from the blood-feeding sand fly Phlebotomus papatasi. The expressed Phlebotomus enzyme was a calcium-dependent ADP and ATPase, with a pH optimum of ∼7, a low KM for ADP and ATP (61 uM and 65 uM, respectively), and a very high substrate turnover rate (3.0 × 104 min−1 and 2.9 × 104 min−1 for ADP and ATP, respectively).

In in vitro platelet aggregometry assays of human platelet rich plasma and PFA-100® analyses of human whole blood under high shear conditions, the Phlebotomus apyrase was a potent inhibitor of ADP induced platelet aggregation. The pharmacokinetics of the apyrase, administered intravenously at a single dose of 12.5 mg/kg, was also investigated in rats. In vivo, the enzyme was distributed and eliminated rapidly, with a half-life of approximately 15 minutes. To examine the efficacy of the apyrase as an anti-thrombotic agent, a rat vascular electrolytic injury model was utilized. Intravenous administration of apyrase at a dose of 12.5 mg/kg showed a statistically significant ∼35% inhibition in occlusive thrombus formation compared to control animals. A clear dose-response relationship was observed, with approximately 22% and 15% inhibition seen at 1.0 and 0.3 mg/kg doses, respectively. No inhibition was achieved at the 0.1 mg/kg dose. The platelet P2Y12 ADP receptor antagonist clopidogrel at 10 mg/kg in this model exhibited ∼41% inhibition of thrombus formation.

Soluble apyrases act through the enzymatic “deletion” of the platelet pro-aggregant ADP, inhibiting platelet aggregation and thrombus formation. Thus, the unique anti-platelet actions of the apyrase enzymes make them particularly attractive in the context of therapeutic agents for the inhibition of platelet-mediated thrombotic disorders.

Characterization of Adenosine Metabolism and Coronary Flow Control by Adenosine in Mouse Heart

J. Weichsel, A. Pexa & A. Deussen

Department of Physiology, Med. Fakultät Carl Gustav Carus, TU Dresden, Germany johnwei@gmx.de

The murine heart preparation has been increasingly used in recent years, because of the opportunity to employ knockout or overexpression models. A further aspect relates to the use of expensive pharmacological tools, which is more economical in this species due to the small organ size. Aims of the present study were the characterization of adenosine metabolism and purine related coronary flow regulation of mouse heart.


C57Bl6 mice weighing 19–25 g were anaesthetized and heparinized (urethane 2mg/g body mass, Liquemin® 500 I.U.). Hearts were isolated and retrogradely perfused via the ascending aorta according to the Langendorff-method providing 37°C warm Krebs-Henseleit-buffer equilibrated with carbogen gas (pO2 680 mm Hg, pCO2 35 mm Hg) at a perfusion pressure of 85 ± 3 mmHg. Heart rate was kept constant by pacing (505 ± 5 bpm) and coronary flow was measured using an Ultrasonic® flow probe. A PVC-balloon was inserted into the left ventricle via the left atrium to allow measurement of ventricular pressure development and dP/dt. A catheter advanced into the right atrium was used for collecting venous effluent perfusate for analysis of adenosine concentration (HPLC technique) and gas partial pressures.


Baseline parameters were (mean ± S.E.M.): flow 9.5 ± 1 ml/min*g, dP/dtmax 4097 ± 243 mmHg/s, coronary venous pO2 160 T 7 mmHg, and adenosine concentration 18 ± 3 nmol/l. The reactive flow overshoot following a 20 s flow stop was 280 ± 9% (n = 10) of baseline flow. Elevation of the coronary arterial adenosine concentration by 80 and 200 nM resulted in marked increases of coronary flow (2.8-, 3.4-fold), an increased contractility (1.5-, 1.3-fold) and elevated venous pO2 (2.3-, 2.7-fold), respectively. Block of adenosine deamination and phosphorylation by EHNA (5 µM) and ITU (10 µM), respectively, augmented adenosine release 8.9-fold. In parallel, coronary flow rose 3.2-fold. In presence of EHNA and ITU, NBTI (10 µM), a specific blocker of equilibrative nucleoside transport, only moderately lowered adenosine release by 18%. Specific block of A1, (DPCPX 100 nM), A2A (ZM-241385 100 nM, SCH-58261 100 nM) and A2B (Alloxazine 10 µM, MRS-1754 200 nM) adenosine receptors did neither change adenosine release nor baseline coronary flow significantly. However, the effect of exogenous adenosine (80 nM, 200 nM) on coronary flow was almost completely blocked by A2A antagonism, partly by A2B antagonism but not by blocking A1 adenosine receptors. During hypoxic perfusion, A2A antagonists lowered coronary flow by approximately 10%.


1) The coronary flow control of mouse heart exhibits a high sensitivity for adenosine. 2) Exogenous adenosine acts mainly via activation of A2A-receptors. 3) Selective blockade of A2A-receptors does not largely affect baseline coronary flow and adenosine release. 4) During hypoxic perfusion endogenous adenosine contributes to flow control. 5) Block of membrane transport of adenosine is rather insensitive to NBTI (NBMPR) as compared to other species.

Characterization of the platelet P2Y12 receptor with the new, selective radiogland [3H]PSB-0413

K.J. Griessmeier, A. El-Tayeb & C.E. Müller

Pharmaceutical Sciences Bonn (PSB), Pharmaceutical Chemistry, Institute of Pharmacy, University of Bonn, Kreuzbergweg 26, 53115 Bonn, Germany kerstin.griessmeier@uni-bonn.de

The P2Y12 ADP receptor is one of the major regulators of hemostasis and thrombosis and the target of antithrombotic thienopyridines and nucleotide analogs. It is a G protein-coupled receptor which upon activation inhibits adenylate cyclase activity [1, 2]. The platelet P2Y12 receptor has been extensively characterized in functional assays [1, 2]. However, characterization on the protein level has been hampered by the lacking of a selective radioligand.

In the present study, we characterized the P2Y12 receptor in human platelet membrane preparations using the newly synthesized enzymatically stable and P2Y12-selective radioligand [3H]PSB-0413 (2-propylthioadenosine-5′-adenylic acid (1,1-dichloro-1-phosphonomethyl-1-phosphonyl) anhydride, AR-C67085MX) [3]. [3H]PSB-0413 showed a high affinity for P2Y12 receptors natively expressed in human platelets (KD = 4.57 ± 0.51 nM) as determined in a saturation binding assay. The kinetically derived KD value was similar (3.69 ± 1.58 nM at r.t. and 4.87 ± 2.81 nM at 4°C, respectively). A membrane preparation of human platelets showed a high expression level of P2Y12 receptors (Bmax = 7.66 ± 0.69 pmol/mg of protein). It was much higher than than the density of P2Y1 receptors (170 fmol/mg protein). A competition assay for the screening of ligands was established and a series of standard agonists and antagonists was evaluated. The rank order of potency for the inhibition of [3H]PSB-0413 binding was (Ki values): 2-methylthio-ADP (4.88 nM) ≫ ADPβS (4.11 µM) > ATPγS (13.2 µM) = ATP (15.7 µM) = ADP (17.6 µM). GTP shift experiments confirmed that 2-methylthio-ADP and ADP were agonists and that ATP and the precursor of [3H]PSB-0413, PSB-0412 bearing a 2-propargyl instead of a 2-propyl substituent (Ki value = 2.41 ± 0.25 nM), were antagonists at human P2Y12 receptors of platelets. Radioligand binding was not inhibited by 100 µM of the following P2 receptor ligands: pyridoxalphosphate-6-azophenyl-2′,4′-disulphonic acid (PPADS), β,γ-methylene-ATP, α,β-methylene-ATP, UDP-glucose, AMP, IDP, UMP, UDP or UTP. The radioligand is selective versus the closely related P2Y receptor subtypes P2Y11 (EC50 = 1.5–8.9 µM) and P2Y13 (IC50 = 213–630 nM). These results indicate that [3H]PSB-0413 specifically labels the P2Y12 receptor. In conclusion, [3H]PSB-0413 is the first selective high-affinity antagonist radioligand for the P2Y12 receptor.

Supported by the Deutsche Forschungsgemeinschaft (GRK-677)

Characterization of the receptor(s) that mediate BzATP responses in rat cerebellar astrocytes

Esmerilda G. Delicado, Luz María G. Carrasquero and Ma. Teresa Miras-Portugal

Departamento de Bioquímica, Facultad de Veterinaria, Universidad Complutense de Madrid, Spain esmeril@vet.ucm.es

Cerebellar astrocyte cultures is one of the experimental models used to study nucleotide signaling in cerebellum. In previous works, we showed that cerebellar astrocytes possess a great diversity of functional P2Y receptors, ADP sensitive receptors, the P2Y1 and/or the P2Y13-like receptors, (see Poster of Carrasquero et al.) and ATP/UTP sensitive receptors, the P2Y2/P2Y4 receptors, mainly coupled to PLC activation. These results contrast to those reported for astrocytes from other brain areas, such as cortex or hippocampus, which also express some ionotropic P2X receptors, including the P2X7 subtype. The P2X7 receptor is an ionotropic P2X receptor, only activated at millimolar ATP concentrations in vivo, which exhibits very peculiar characteristics. It was thought to be exclusive for cells of haematopoietic lineage, where its role in cellular toxicity and inflammatory process is well established. Recently, it has been also detected at the nervous system, where has resulted to be an “intriguing” receptor and far from induce cellular lysis or apoptosis, could exert other actions, such as differentiation, glutamate release, etc. Basing on these findings, we decided to explore their presence in cerebellar astrocytes using BzATP, described as the most specific agonist for this receptor subtype. We have found that cerebellar astrocytes displayed several BzATP responses, which are quite different to those elicited by other nucleotides, which would be mediated by P2X7-like receptors: i) BzATP-induced calcium responses were sustained, and not transient, as those obtained with the metabotropic agonists 2MeSADP and UTP, ii) Among the nucleotides BzATP produced the maximal ERK activation, iii) BzATP also produced long-term effects, inducing morphological changes, that leads to differentiation. Moreover, the presence of mRNA codifying of P2X7 receptor and the protein was also confirmed. Although P2X4 subunits have been also detected, their functionality remains unclear.

The characterization of BzATP calcium responses was made by calcium imaging using fura-2. Most of tested cells (80%) were sensitive to BzATP stimulations eliciting calcium responses, which are biphasic. They exhibited initial transients with an EC50 value of 12.2 + 1.6 µM, followed by a sustained responses, that were not desensitized. But, surprisingly, when BzATP challenges were applied in the absence of extracellular calcium, the initial transients were maintained, whereas as expected the sustained responses were completely abolished, which clearly indicated that the transients resulted from intracellular calcium mobilization. The same occurred by preincubation with 1 µM Brilliant Blue G, concentration at which this antagonist selectively acts on P2X7 receptors and stabilize the inactive form of the receptor, the sustained responses were abolished, but the transient was also observed. BzATP calcium responses were not affected by cross-desensitization with UTP or 2MeSADP, when were consecutively applied, though each agonist was able to desensitize itself, indicating that they are acting on different receptors. These data suggest that BzATP possess specific receptors, P2X7-like receptor type, in cerebellar astrocytes, which are also coupled to PLC activation.

Chronic cyclosporine treatment increases total homocysteine levels and decreases adenine-nucleotides hydrolysis in rat blood serum

Ana Elisa Böhmer1, Liz Marina B. P. Brum2, Carolina G. Souza1, Jean P. Oses1, Mariana Streit1, Ricardo S. Bruch2, João J.F. Sarkis1, Luis V. Portela1, Diogo O. Souza1

1Departamento de Bioquímica, Instituto de Ciências Básicas da Saúde, Universidade Federal do Rio Grande do Sul, Rua, Porto Alegre, RS, Brazil. 2Instituto de Cardiologia do Rio Grande do Sul, FUC, Porto Alegre, RS, Brazil. anaelisab@uol.com.br

Cyclosporine (CsA) is a potent immunosuppressant agent that has been extensively used in transplanted patients. However, CsA administration is accompanied by a variety of adverse effects being allograft vasculopathy the major cause of morbidity and mortality among renal and cardiac transplant recipients. The mechanisms involved in the vascular injury and the putative participation of CsA are poorly understood. CsA administration might affect homocysteine (Hcy) serum concentrations, a sulfur containing amino acid recognized as an independent risk factor for atherosclerosis and venous thromboembolism. Excessive production of free radicals, increase on platelet adhesiveness and decrease on tissue and plasmatic adenosine levels contribute to the vascular damage caused by Hcy. Extracellular adenosine has important benefic effects on the vascular system, avoiding thrombus formation and circulatory problems through the regulation of platelet aggregation, vasodilatation, coronary blood flow and inflammation. Thus, control of the ratio of adenine nucleotide/adenosine in the extracellular space by ecto-nucleotidases activities is one of the important steps for the maintenance of vascular homeostasis.

In this work, we examined the effects of chronic CsA administration on total serum Hcy levels, on the adeninenucleotides hydrolysis (ATP, ADP, AMP) through serum ecto-nucleotidases activities and its putative association with vascular disturbance in adult male Wistar rats. Animals were daily administered with CsA 5 mg/kg, 15 mg/kg or vehicle via gastric gavage during 8 weeks. Total Hcy concentrations in serum was measured using a commercial MEIA kit (Abbott, USA) and ATP, ADP and AMP hydrolysis were evaluated through phosphate release, using the method described by Oses et al. [1]. CsA induced a significant increase on total Hcy concentration and a decrease in ATP, ADP and AMP hydrolysis. The inhibition of nucleotides hydrolysis was negatively correlated with total Hcy levels and positively with uric acid levels. Increased platelets number and fibrinogen levels, involved in coagulation pathway, were also observed.

Altogether, these results indicate that chronic CsA treatment affects the homeostasis of vascular system by increasing platelets, fibrinogen and serum levels of total Hcy. Moreover, we postulated that by inhibiting adenine nucleotides hydrolysis, CsA might disrupt the equilibrium between adenine nucleotides/adenosine levels. Finally, these alterations could be implicated in the vascular complications reported in patients under CsA therapy. (Supported by CNPq Brazil, FAPERGS).

Cloning and identification of the liver canalicular ECTO-ATPase as NTPDase8

M. Fausther1, J. Lecka1, F. Kukulski1, J. Pelletier1, S.A. Lévesque1, H. Zimmermann2, J.A. Dranoff3 and J. Sévigny1

1Centre de Recherche en Rhumatologie et Immunologie, Université Laval, Québec, Canada

2Biocenter, J.W. Goethe-University, AK Neurochemistry, Frankfurt am Main, Germany

3Section of Digestive Diseases, Yale University School of Medicine, New Haven, CT, USA michel.fausther@crchul.ulaval.ca


Liver cells express various ecto-nucleotidases including CD39/ecto-nucleoside triphosphate diphosphohydrolase-1 (NTPDase1), CD39L1/NTPDase2 and CD73/ecto-5′-nucleotidase to regulate nucleotide and nucleoside levels at the cell surface. Since the these molecules control a number of critical cellular functions like cell volume autoregulation and ionic secretion, the biological activity of ecto-nucleotidases appears essential for liver homeostasis maintenance. Interestingly, the liver has one of the highest ecto-nucleotidase activity among tissues. In the liver, NTPDase1 is expressed by Kupffer cells and vascular endothelial cells whereas NTPDase2 is produced by portal fibroblasts and activated hepatic stellate cells. However, their combined activity appears much lower than the total ecto-nucleotidase activity observed in liver tissue. Histochemical studies showed that the high liver ecto-nucleotidase activity is mainly associated with the canalicular domain of hepatocytes and has the general characteristics of an NTPDase activity. We have recently reported, in mouse, the cloning of a novel ectonucleotidase, NTPDase8 whose cDNA is highly expressed in liver. The Aim of this study was to identify the main ecto-nucleotidase(s) in the liver, particularly the one(s) expressed in the canalicular domain of hepatocytes.


NTPDase8 cDNA was cloned from human and rat liver tissues and inserted in pcDNA3 expression vector. Constructs were used for transfection assays in COS-7 cell line. NTPDase8 was purified from rat liver by a 3-step chromatography. The biochemical characterization was performed with protein extracts from NTPDase8-transfected cells or the purified protein. Antibodies against rat NTPDase8 and CD73 were generated and used to examine the expression pattern of these proteins, by Western blot and immunofluorescence.


Activity assays on protein extracts (from COS-7 cells expressing rat or human NTPDase8) and purified rat NTPDase8 regarding substrate specificity, ion requirement and sodium azide inhibition showed that the enzyme displayed similar biochemical properties as previously described for the purified porcine canalicular ecto-ATPase. The analyses of the primary structures showed high resemblance between rat and human NTPDase8 when compared to porcine ecto-ATPase. The specificity of polyclonal antibodies against rat NTPDase8 and CD73 was confirmed by immunoblot and immunocytochemistry, using transfected cells. By western blot, the anti-rat NTPDase8 detected a protein with a molecular weight of 75 kDa with the highest expression in liver and also some expression in kidney and jejunum. In liver sections, NTPDase8 was expressed in the canalicular membrane domain of hepatocytes and co-localized with the canalicular marker MRP-2. Immunostaining for CD73 showed that this enzyme was also present in the bile canaliculi.


In this work, we report the cloning of NTPDase8 in human and rat species, its identification as the canalicular ecto-ATPase and demonstrate that it represents the major ecto-nucleotidase activity of the liver. In addition, the canalicular localization of both NTPDase8 and CD73 suggest their potential involvement in biliary function.

Cloning, expression and characterization of the Torpedo marmorata E-NTPDase1

Mireia Martín-Satué1, Benjamín Torrejón-Escribano2, Inma Gómez de Aranda1, Antonio Felipe3, Marc Elías1, Jordi Marsal1, Joan Blasi1, and Carles Solsona1

1Cellular and Molecular Neurobiology laboratory, Dep. of Therapeuthics and Experimental Pathology, Medicine School, Bellvitge Campus, University of Barcelona, Spain

2Scientific and Technical Services, Bellvitge Campus, University of Barcelona, Spain

3Department of Biochemistry and Molecular Biology, University of Barcelona, Spain martinsatue@ub.edu

Ecto-nucleoside triphosphate diphosphohydrolases (E-NTPDases) constitute a family of enzymes that hydrolyse both ATP and ADP to AMP, differing in their substrate preferences and tissue distribution. They are broadly expressed enzymes, mainly membrane bound, playing a relevant role in several physiological processes. An E-NTPDase activity in electric organ membranes of Torpedo fishes was previously identified (1) and we have been using these membranes as a source of E-NTPDase activity to study the effect of new possible inhibitors (2,3,4). Enzymology studies pointed to the classification of this enzyme into the class-1 E-NTPDases, since it shows nearly equal preference for ATP and ADP (5). We report here the cloning of its full-length cDNA and its characterization.

The clone, obtained by means of RACE (Rapid Amplification of cDNA Ends)-PCR technique, contains an open reading frame of 1506 bp and codes for a 502 aminoacids-protein that exhibits high homology with other E-NTPDases1 previously identified, including those of zebrafish or Xenopus but also human, rat and mouse. Topology analyses revealed the existence of two transmembrane regions, two short cytoplasmic tails and a long extracellular domain containing the five highly conserved apyrase regions. Gene expression studies revealed that this gene is expressed in all the Torpedo tissues analysed. We have also characterized the expression of the recombinant protein expressed in COS and HeLa cells.

This work was supported by grants of Ministerio de Educación y Ciencia from the Spanish Governement (C.Solsona, project no. SAF2005-00736) and Comisió de Recerca del Campus de Bellvitge (M. Martín-Satué, project no. ACESB/13) from the University of Barcelone, Spain.

Co-localization and functional cross-talk between A1 and P2Y1 purine receptors in the brain

I. Tonazzini1, M.L. Trincavelli1, L.H. Bergersen2, J. Storm-Mathisen2, M.P. Abbracchio3, C. Martini1

1Dept. of Psychiatry Neurobiology Pharmacology and Biotechnology, University of Pisa, Italy. 2Dept. of Anatomy and Centre for Molecular Biology and Neuroscience CMBN, University of Oslo, Norway. 3Dept. Pharmacological Science, Unviversity of Milan, Italy tonazzini@farm.unipi.it

Adenosine and ATP are released from glia and neuron cells and are involved in a wide variety of both physiological and pathological processes by their interaction with specific receptors classified as P1 and P2, respectively (Abbracchio and Burnstock, 1998). Whereas the neuroprotective role of adenosine analogues active at the adenosine A1 receptor subtype (A1AR) have been known for several years (Wardas, 2002), the role of purinergic P2Y1 receptor (P2Y1R), enough if not fully characterized, indicate important pathological functions/involvements: purine-mediated induction and maintenance of astrogliosis can be considered as a first response to limit the loss of neuronal tissue after insults (Franke and Illes, 2006). Recent data suggest the existence of a heteromeric complex between A1ARs and P2Y1 purinergic receptors with novel pharmacological and functional properties (Yoshioka and Nakata., 2004). As evidenced by G-protein coupled receptors heterodimerization studies, an important role in the control of receptor signalling and regulation processes has been demonstrated (Maggio et al., 2005).

In the present work we investigated the co-localization of A1 and P2Y1 receptors on rat brain tissues and their heterodimerization and functional cross-talk in astroglial cells (ADF).

Immunogold-electron microscopy quantification was used to study the cellular localization of A1 and P2Y1 receptors on rat hippocampus which is considered as a damage sensitive brain area. The results demonstrated a significantly high expression of both receptors on synaptic and astroglia membranes. Since glial elements represented the cell population in which the higher A1/P2Y1 co-localization occurred, an astroglial cell line was used to investigate the receptor functional cross-talk. A1AR and P2Y1R expression in ADF cells was demonstrated by immunoblotting experiments; moreover, in co-immunoprecipitation assay we showed that these receptors constituted an heteromeric complex in basal conditions.

A1AR/P2Y1R functional interaction was investigated evaluating the heterologous regulation of A1AR-cAMP pathway induced by the P2Y1-agonist, 2-MeSADP.

In control ADF cells, A1AR agonist, CHA, was able to inhibit, in concentration dependent manner, the production of cAMP (EC50 = 0,9 nM). ADF cell exposure to 2-MeSADP inhibited A1AR functional response in time dependent manner with a maximal effect after 30 minutes. Moreover, a significant right shift of CHA dose-response curve occurred. These data demonstrated P2Y1R induced heterologus A1AR desensitisation in ADF cells.

To clarify the cellular co-localization, the interaction and the function of heterodimerization of A1 and P2Y1 receptors and the possible alterations of P1–P2 receptor regulation processes during pathological conditions may help to clarify the patho-physiological functions of ATP and adenosine in CNS.

Comparison of NTPDase 1, 2 and 3 after refolding from bacterial inclusion bodies

Matthias Krause & Norbert Sträter

From the Biotechnologisch-Biomedizinisches Zentrum, Fakultät fü r Chemie und Mineralogie der Universität Leipzig, Deutscher Platz 5, 04103 Leipzig, Germany matthias.krause@bbz.uni-leipzig.de

The ecto-Nucleoside triphosphate diphosphohydrolases or NTPDases are the enzymes responsible for signal termination and conversion in purinergic signalling. NTPDase1, 2 and 3 are localized on the cell surface, anchored to the membrane by two transmembrane helices. They catalyze the sequential removal of γ- and β- phosphate from ATP, ADP and other nucleotides.

Due to their involvement in physiological processes like blood clotting and pain perception, but also in certain types of cancer, they are now considered as potential new drug targets or drugs themselves. Soluble forms from mammalian cell culture have been shown to inhibit platelet aggregation in vivo. However, mammalian production of large amounts of therapeutic proteins is expensive and bears the risk of viral contamination. Therefore and for structural characterization we established an E. coli expression system for insoluble production of the extracellular domains of three NTPDases and optimized the in vitro refolding. The refolded proteins have been characterized in several aspects: substrate specificity, KM values for ATP and ADP, pH dependance and metal cofactor activation. Similarities and differences in the refolding behaviour and in enzymatic properties among the recombinant enzymes and in comparison to their wildtype counterparts will be discussed.

Our expression system allows for large scale production of active protein for further characterization and clinical application. X-ray structure determination of the refolded proteins will provide valuable insights in the rational design of medically relevant inhibitors.

Complex Kinetics of P2X7-Dependent Single Channel Currents

T Riedel1, I Lozinsky1, G. Schmalzing2 and F Markwardt1

1Julius-Bernstein-Institut für Physiologie, MLU Halle; 2Pharmakologisches Institut, RWTH Aachen fritz.markwardt@medizin.uni-halle.de

P2X receptors are ATP-gated ion channels permeable to small inorganic cations. The P2X7 receptor has several peculiar characteristics compared to the other members of the P2X receptor family. The time course of activation and deactivation is dependent on species, number and duration of ATP applications. The initial part of activation is supposed to be so fast that it could not be investigated by conventional whole cell voltage clamp methods. Therefore, we expressed human P2X7 receptors heterologously in Xenopus oocytes and investigated the kinetics of activation and deactivation of P2X7-dependent single channel currents by means of the patch clamp method in the outside-out configuration. A combination of the U-tube technique and a piezo-driven liquid filament switch technique enabled the application and withdrawal of different concentrations of the agonist ATP4− within less then 1 ms. Two different types of kinetics were observed and approximated by single exponential functions. A low-noise patch current component activated and deactivated with time constants of about 300 ms. Its amplitude was dependent on [ATP4−] and the seal resistance but independent of the expression of human P2X7 receptors. It was attributed to non-specific effects of ATP on the seal resistance. Extracellular application of ATP evoked slightly inward rectifying single-channel currents with a conductance of about 10 pS and mean open times of about 5 ms at negative membrane potentials. The mean closed times but not the mean open times were dependent on the activating ATP concentration. High ATP concentrations increase the probability of the additional opening of channels with a conductance of about 14 pS and mean open times of about 20 ms. At positive membrane voltages, the mean open time was increased to about 15 ms. The channels activated with activation and deactivation time constants of about 15 ms. The channel activation was only little dependent on the agonist concentration. Activation and deactivation were not continuously dependent on the membrane voltage. We conclude that binding of ATP4− is not the only rate limiting step in the activation of P2X7 channels and that the ATP4− binding site is not located in the electrical field of the membrane.

This work was supported by the Deutsche Forschungsgemeinschaft (Ma 1581/12-1) and the Roux-programme of the ML university (FKZ 13/07).

Conduction of purine-induced responses in intracrebral arterioles: relative roles of endothelium and smooth muscle cells

Al C. Ngai1, Thien-Son K. Nguyen1, H. Richard Winn2

1Department of Neurological Surgery, University of Washington School of Medicine, Seattle; 2Department of Neurological Surgery, Mount Sinai School of Medicine, New York richard.winn@mountsinai.org

Purines such as adenosine (Ado) and ATP elicit conducted dilation and/or constriction in cerebral arterioles. In this study, we evaluated the roles of the endothelium and smooth muscle in the conduction of vasomotor responses induced by purines along intracerebral arterioles. Penetrating intracerebral arterioles were isolated from Sprague Dawley rats, and cannulated with a concentric micropipette system. Upon pressurization to 60 mm Hg, the vessels developed spontaneous tone, contracting to 67 ± 1% of passive diameter (77 ± 4 µm, n = 9). The experimental protocol consisted of pulse-application of vasoactive agents (Ado or ATP) via micropipettes, onto a short segment of the arteriole. This induced both a direct local response, and a secondary conducted response that spread longitudinally along the vessel. Ado is an endothelium-independent dilating agent, whereas ATP-induced dilation is endothelium dependent. Pressure-pulse ejection of Ado (10 mM in pipette) elicited a local 14% dilation, and a 4% dilation at a site 500 µm from the ejection site, whereas ATP (10 mM) microapplication evoked biphasic local (18% constriction, 9% dilation) and conducted (11% constriction, 8% dilation) responses. Endothelial cells within an ∼100 µm segment (positioned between the application and observation sites) were injured by a luminal light-dye (L-D) technique using luminal Na Fluorescein. Alternatively, vascular smooth muscle (VSM) cells were selectively damaged by adventitial L-D treatment. Endothelial injury did not significantly affect the conducted dilation responses to Ado, but virtually eliminated the conducted responses (both dilation and constriction) to ATP. VSM injury abolished conducted dilation responses to Ado, but did not affect ATP-dilation. However, VSM injury attenuated the remote constriction responses to ATP microapplication. Our data suggest that the endothelium is the predominant pathway for the conduction of ATP-induced dilation, whereas the smooth muscle layer primarily mediates the conduction of Ado-induced dilation. Both the endothelium and smooth muscle appear to be involved in the propagation of ATP-induced constriction.

Funded by NIH Grant NS-21076 and AHA Grant 0255703N.

Constitutively Active Mutants of the Human Adenosine A2B Receptor Reveal Inverse Agonism

Qilan Li1, Kai Ye1, Clara C. Blad1, Hans den Dulk2, Jaap Brouwer2, Ad P. IJzerman1, Margot Beukers1

1Medicinal Chemistry (LACDR) and 2Molecular Genetics (LIC), Gorlaeus Laboratories, Einsteinweg 55, 2300 RA, Leiden, the Netherlands. Q.li@chem.leidenuniv.nl

The human adenosine A2B receptor belongs to class A G-protein-coupled receptors. In our previous work, constitutively active mutant (CAM) human adenosine A2B receptors were identified from a random mutation bank based on their ability to grow in histidine-deficient medium [1]. In the current study, three known A2B receptor antagonists, MRS1706, ZM241385 and DPCPX were tested on wild-type and 9 CAM A2B receptors with different levels of constitutive activity in a yeast growth assay.

The compounds’ ability to antagonize the agonist NECA on the wild-type receptor was assessed first. The pA2 values of MRS1706, ZM241385 and DPCPX were derived from a Schild analysis, and found to be in good agreement with available literature.

Since the wild-type receptor lacks constitutive activity, we employed CAM receptors to investigate whether these antagonists could be reclassified as inverse agonists. All three compounds turned out to be inverse agonists for the adenosine A2B receptor as they were able to fully reverse the basal activity of the 4 low-level CAM A2B receptors: T42A, F84S, F84L and F84L/S95G. The basal activity of 3 medium-level CAM A2B receptors, N36S/T42A, T42A/ V54A and N36S/T42A/T66A, was partially inhibited.

Figure 1

Inverse agonism of ZM241385 on CAM receptors with varying levels of constitutive activity.

We also discovered 2 highly constitutively active or locked mutants, A18T/A23V/C83Y/A106V/R112S and Q21-4L/I230N/V240M/V250M/N254Y/T257S/K269stop (a truncated receptor). Their basal activity could not be reversed by any of the three compounds. The inverse agonistic properties of ZM241385 on the T42A, N36S/T42A/T66A and the truncated receptor are shown in figure 1 as a typical example.

The rank order of potency of the compounds matched the rank order obtained in the antagonist assay on the wildtype receptor, that is ZM241385 = MRS1706 > DPCPX. However, ZM241385 was the strongest inverse agonist as demonstrated by the rank order of intrinsic activities ZM241385 > MRS1706 = DPCPX.

In conclusion, this study is the first to describe inverse agonism on the human adenosine A2B receptor. Moreover, the use of receptor mutants with varying levels of constitutive activity enabled us to appreciate differences in intrinsic activity of the inverse agonists.

Contractile Effects of Adenosine, Coronary Flow and Perfusion Pressure in Murine Myocardium

Laura Willems and John P. Headrick

Heart Foundation Research Centre, Griffith University Southport, QLD 4217, Australia j.headrick@griffith.edu.au

There is mixed evidence adenosine receptors (ARs) may directly enhance myocardial contractility, though this remains contentious. Identifying such effects is important in clarifying physiological and protective functions of adenosine. We assessed inotropic actions of adenosine (50 µM) and selective AR activation with 100 nM N6-cyclohexyladenosine (CHA; A1AR agonist), 25 nM 2-[p-(2-carboxyethyl) phenethylamino]-5′-N-ethylcarboxamidoadenosine (CGS-21680; A2AAR agonist), and 100 nM 2-chloro-N6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (Cl-IB-MECA; A3AR agonist) in C57/BL/6J mouse hearts paced at 420 beats/min, and perfused at constant-pressure, constant-flow, or under conditions of stable flow and pressure (achieved with nitroprusside-mediated vasodilatation in constant-flow hearts, prior to AR activation). Adenosine and CGS-21680 induced significant (albeit modest) positive inotropy in constant-pressure perfused hearts (up to 10 mmHg elevations in systolic force). Inotropic effects of adenosine and CGS-21680 paralleled coronary vasodilatation (with up to 10 ml/min/g elevations in flow). Neither CHA or Cl-IB-MECA altered force or flow. Under conditions of constant flow, adenosine and CGS-21680 reduced systolic pressure in parallel with coronary perfusion pressure. When changes in coronary flow and perfusion pressure were prevented (eliminating potential Gregg-related effects), CGS-21680 no longer modified contractile force. However, adenosine itself still significantly enhanced systolic pressure by up to 10vmmHg, independently of changes in coronary flow, perfusion pressure, and heart rate. Relations between flow, perfusion pressure, and ventricular performance evidence significant Gregg-related effects in murine myocardium — ventricular systolic pressure increases transiently by ∼1 mmHg per ml/min/g change in flow during the initial 1–2 min of hyperemia, and in a sustained manner by ∼1 mmHg per mmHg change in coronary perfusion pressure. These effects contribute to inotropic effects of AR agonists when coronary flow or pressure are uncontrolled. In summary, we find no evidence of direct A1 or A3AR-mediated changes in contractility in intact myocardium. Inotropic actions of A2AAR agonism are indirect, involving Gregg-related effects. Despite no A1, A2A or A3AR-mediated effects, the endogenous agonist adenosine exerts a modest inotropic action independently of flow and perfusion pressure. The mechanistic basis of this response remains to be identified.

Contribution of PKC to the desensitisation of relaxation to luminally-perfused purinoceptor agonists in rat small mesenteric arteries

Polly Winter & Kim. A. Dora

Department of Pharmacy & Pharmacology, University of Bath, Bath BA2 7AY, UK pr9pw@bath.ac.uk

Protein kinase C (PKC) contributes to the desensitisation of the P2Y1 purinoceptor in response to prolonged treatment with the non-hydrolyzable analogue of ADP, adenosine 5′-[β-thio]diphosphate (ADPβS) in isolated blood platelets [1]. The aim of this study was to ascertain the extent of PKC modulation of the vasodilatation of small rat mesenteric arteries in response to luminal perfusion of ATP, the non-hydrolysable analogue of ATP, adenosine-5′-(3-thiotriphosphate) ATPγS, and ADPβS.

Segments of third order mesenteric arteries were isolated from male Wistar rats (200–250 g), killed by schedule 1 methods of the Animals Scientific Procedures Act 1986 (UK). Arteries were cannulated, mounted in a pressure myograph (Danish Myotechnology), pressurized to 50 mmHg and perfused with MOPS at 37°C. Following precontraction with phenylephrine a multi-channel syringe pump was used to perfuse agonists through the lumen at 90 µl.min−1. Once the peak relaxation response was established the time course of desensitisation of the relaxation response was assessed over a 2 min period. Values are means ± s.e.mean of the percentage relaxation to maximum diameter.

Luminal perfusion of ATP (3 µM) and ATPγS (3 µM) resulted in peak vasodilatation responses of 91.3 ± 1.5% and 80.5 ± 4.0% respectively. These levels of relaxation were maintained throughout the 2 min post-peak period (3 µM ATP, 96.4 ± 0.8%; 3 µM ATPγS, 78.4 ± 12.1%) and were unchanged following luminal incubation (15 min) with the selective inhibitor of PKC, bisindolylmaleimide 1 (BIS 1 µM) (3 µM ATP, 94.6 ± 1.4%; 3 µM ATPγS, 88.8 ± 1.5%). In contrast, luminal perfusion of ADPβS (1 µM) and ATP (1 µM) resulted in relaxation responses of 70.0 ± 6.6% and 40.5 ± 8.2% respectively that decayed over the 2 min post-peak period (1 µM ADPβS, 41.5 ± 6.1%; 1 µM ATP, 28.9 ± 9.3%). Prior luminal incubation with BIS increased the peak relaxation response to ADPβS (1 µM) (94.8 ± 1.0%, P < 0.05; students paired t-test) and ATP (1 µM) (70.4 ± 14.5%). The decay of the responses observed in response to these two agonists was also significantly reversed (1 2MADPβS, 89.1 ± 6.6%; 1 µM ATP, 67.9 ± 13.8%, P < 0.01, students paired t-test).

PKC has been found to contribute to the desensitisation of the relaxation response to ADPβS (1 µM) and ATP (1 µM) supporting a role for PKC in the desensitisation of P2Y1 receptors within the endothelium of isolated rat mesenteric arteries. Activation of P2Y2 or P2Y4 receptors with ATPγS (3 2M) did not evoke a response that decayed over time. Variation in the downstream pathways beyond each P2Y purinoceptor subtype may therefore exist.

This work was kindly supported by the British Heart Foundation

Contributions of P2Y and P2X receptors to nucleotide-induced signaling in macrophages revealed by single-cell Ca2+ measurements and knockout mice

Adriana del Rey1, Vijay Renigunta1, Alexander H. Dalpke2, Jens Leipziger3, Joana E. Matos3, Bernard Robaye4, Marylou Zuzarte1, Annemieke Kavelaars5 and Peter J. Hanley1

1Institute of Physiology, Marburg University, Germany; 2Department of Hygiene & Medical Microbiology, Heidelberg University, Germany; 3Institute of Physiology & Biophysics, Aarhus University, Denmark; 4Institute of Interdisciplinary Research, Université Libre de Bruxelles, Belgium; and 5Psychoneuroimmunology, University Medical Center Utrecht, The Netherlands hanley@mailer.uni-marburg.de

Antigen-presenting cells express two families of nucleotide receptors: ATP-gated cation channels (P2X receptors) and P2Y receptors, a subset of the G protein-coupled receptor superfamily. To identify the receptors mediating the Ca2+ responses of resident peritoneal macrophages to extracellular UTP, ATP and UDP, we performed Ca2+ measurements on single cells isolated from various knockout mice. To obviate the confounding effects of extracellular Ca2+ sources, experiments were performed in Ca2+-free buffer. Note that when ATP is applied to a macrophage in the presence of Ca2+, the observed increase in cytosolic [Ca2+] ([Ca2+]i) can be attributed to endoplasmic reticular (ER) Ca2+ release, Ca2+ influx through Ca2+ release-activated Ca2+ (CRAC) channels or Ca2+ influx via P2X receptors (Figure 1). We found that application of UTP (or ATP), under Ca2+-free conditions, to macrophages isolated from wild-type mice induced a transient (lasting 1 min) increase in [Ca2+]i with half-maximal response at concentrations <1 µM. However, macrophages isolated from P2Y2/P2Y4 double knockout mice did not respond to either UTP or ATP. Furthermore, the nucleotide-induced Ca2+ response of macrophages isolated from P2Y4−/− mice was normal whereas cells from P2Y2−/− mice were completely unresponsive. Thus, P2Y2 is the sole receptor which couples extracellular UTP or ATP to phospholipase C-β (PLC-β) activity and ER Ca2+ release. In contrast to UTP, UDP (purified with hexokinase) was an essentially ineffective agonist in both P2Y2-deficient and wild-type macrophages, suggesting that the P2Y6 receptor is weakly expressed. In P2Y2−/− macrophages, though, the Ca2+ responses to ‘pure’ P2X receptor activation could be studied by applying ATP in the presence of extracellular Ca2+.

Figure 1

Schematic diagram of Ca2+ signaling in a mouse macrophage induced by extracellular nucleotides.

Conventional protein kinase C isoforms mediate human sperm capacitation induced by A1 adenosine receptor agonist

Lavinia Liguori, Rossana Palazzo, Ilaria Bellezza & Alba Minelli

Dipartimento di Medicina Sperimentale Scienze Biochimiche, Sezione Biochimica Cellulare, Università di Perugia, via del Giochetto, 06123, Perugia, Italia. albaminelli@virgilio.it

Mammalian ejaculated spermatozoa are infertile and the acquisition of the fertilizing capacity is obtained via a sequential two-steps activation process, named capacitation and acrosome reaction. Capacitation, regarded as a signal transduction receptor-mediated phenomenon, is an emerging concept (1,2) and our data, showing that the stimulation of A1AR triggers a cascade of signalling events that leads to capacitation, support this concept. The constant presence of A1AR on spermatozoa suggested a functional role (3), later identified either in human or in murine species as capacitative (4,5). Murine spermatozoa, lacking the A1AR, showed a delayed capacitation indicating that the A1AR modulates the velocity of the process without affecting the number of capacitated cells (4). We have already shown that the signalling initiated by A1AR proceeds via PLC and IP3 release (5). The comparative analysis of the effects of CCPA and FCSu during a 3.5 h incubation shows that CCPA is as effective as FCSu, a known inducer of capacitation, although it does not stimulate O2· production and does not cause an increase of the phosphorylation either of the tyrosine residues or of the Thr-Glu-Tyr-motif suggesting that when the A1AR is activated, its signalling proceeds via kinase-induced events that target proteins different from those obseved in the presence of FCSu. Treatment of human spermatozoa with different protein kinases inhibitors shows that receptor type PTK, PKA, and PI3K are not involved in the CCPA-mediated response whereas PKC is responsible for the capacitative effect. All the PKC inhibitors tested, acting with different mechanism and on different PKC isoforms, caused a significant decrease of CCPA-induced capacitation, indicating that all these PKCs are activated via A1AR. This result was not unexpected after the reported involvement of PLC in the A1AR signalling (5). Kinexus analysis of the phosphorylation levels of PKC isoforms after CCPA stimulation shows the activation of the alpha isoform and disactivation of the gamma. PKC alpha isoform has been related to alteration of cell permeability whereas PKCgamma isoform has been linked to Gβ phosphorylation thereby enhancing the potency of G βγ to stimulate adenylyl cyclase activity. Our results are consistent with the increased permeability occurring during the capacitative event and with the shift to Gαi inhibitory signalling induced by A1AR activation.

Conversion of Nucleoside-Based A3 Adenosine Receptor Agonists into Potent and Selective Antagonists: Overcoming the Problem of Species Differences

Kenneth A. Jacobson1, Bhalchandra V. Joshi1, Athena Klutz1, Soo-Kyung Kim1, Hyuk Woo Lee2, Hea Ok Kim2, Lak Shin Jeong2 and Zhan-Guo Gao1

1Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA.

2Laboratory of Medicinal Chemistry, College of Pharmacy, Ewha Womans University, Seoul 120–750, Korea kajacobs@helix.nih.gov

We have systematically probed the effects of base and ribose substitution of adenosine on both the affinity and intrinsic efficacy at the A3 adenosine receptor (AR). The abilities of a nucleoside to bind to and to activate the receptor are structurally distinct. Based on this observation we designed new nucleoside antagonists of the A3AR, which bind potently and selectively, but do not activate the receptor. The goal is to convert agonists that are A3AR-selective in both human and rat into species independent A3AR antagonists. Previously, the spirolactam nucleoside derivative MRS1292 was designed as a cross-species A3AR antagonist, based on chemical constraint of the ribose moiety [1]. This derivative was shown to lower intraocular pressure in the mouse, supporting the use of A3 antagonists in the treatment of glaucoma [2]. The highly selective agonists of the A3AR, Cl-IB-MECA (2-chloro-N6-(3-iodobenzyl)-5′-N-methylcarboxamidoadenosine) and its 4′-thio analogue, were successfully converted into selective antagonists simply by appending a second N-methyl group on the 5′-uronamide position [3]. The 2-chloro-5′-(N,N-dimethyl)uronamido analogues bound to, but did not activate the human A3AR, with Ki values of 29 nM (4′-O, MRS3771) and 15 nM (4′-S, LJ1256), showing >100-fold selectivity over A1, A2A, and A2BARs. Competitive antagonism was demonstrated by Schild analysis. We used rhodopsin-based, dynamic molecular modeling of the A3AR and ligand docking to characterize the putative ligand-binding region and identified distinct amino acid residues related to binding of ligands, such as agonist Cl-IB-MECA and antagonist MRS3771, and activation. In general, removing H-bonding ability or introducing steric rigidity in the region of the ribose 5′-uronamide lowers efficacy at the A3AR, consistent with predictions derived from modeling. Open image in new window

Cooperation between Gq and Gs-linked signaling pathways in regulation of IL-4 by adenosine in mast cells p]Sergey Ryzhov, Anna E. Goldstein, Italo Biaggioni and Igor Feoktistov p]Vanderbilt University, Nashville, TN 37232, USA. igor.feoktistov@vanderbilt.edu

Adenosine provokes bronchoconstriction in asthmatics through acute activation of mast cells and plays an important role in chronic inflammation by upregulating Th2 cytokines in mast cells, thus promoting IgE synthesis by B-lymphocytes. This effect is mediated by A2breceptors because neither the selective A2a agonist CGS21680 nor the selective A3 agonist IB-MECA reproduced it, and the selective A2b antagonist IPDX prevented it. To get insight into molecular mechanisms underlying this unique role of A2b adenosine receptors in promoting Th2 inflammation, we studied intracellular pathways linking this receptor to upregulation of IL-4 in HMC-1 mast cells. A2b receptors stimulate adenylate cyclase via coupling to Gs proteins, and stimulate phospholipase Cβ through coupling to Gq proteins. Inhibition of phospholipase Cβ with U73122 completely blocked adenosine-induced IL-4 secretion. The protein kinase inhibitor Ro-32-0432 had no effect on A2b-mediated, but inhibited PMA-stimulated IL-4 secretion. In contrast, chelation of intracellular Ca2+ inhibited both adenosine and ionomycin-induced IL-4 secretion. This Ca2+-sensitive pathway most likely includes calcineurin and NFAT, because adenosine-induced IL-4 secretion was blocked with cyclosporine A or 11R-VIVIT peptide. Gs-linked pathways also play a role in the A2b-dependent stimulation of IL-4 secretion; inhibition of adenylate cyclase with 2′,5′-dideoxyadenosine, or protein kinase A with Rp-cAMP or H-89, attenuated adenosine-dependent IL-4 secretion. Although stimulation of adenylate cyclase with forskolin did not increase IL-4 secretion on its own, it potentiated the effect of Pasteurella multocida toxin by 2-fold and ionomycin by 3-fold. Both forskolin and adenosine upregulated NFATc1 expression. We conclude that A2b receptors upregulate IL-4 through Gq signaling that is potentiated via crosstalk with Gs-coupled pathways. Thus, our data explain the necessity and underscore the importance of dual coupling of A2b receptors to Gs/Gq proteins with concurrent stimulation of diverse intracellular pathways for adenosine-dependent regulation of IL-4 production in human mast cells.

Cysteine scanning mutagenesis of the extracellular loop region 286–329 of the Human P2X1 receptor.

Jonathan Roberts* and Richard J. Evans p]Department of Cell Physiology 2 Pharmacology, University of Leicester, Leicester, LE1 9HN, UK; *jar20@le.ac.uk p]As P2X receptors show little homology to other ATP binding proteins that have been crystallized models of agonist binding to P2X receptors have been developed from analysis of site-directed mutagenesis studies. Alanine Communications 161 Springer mutagenesis of conserved amino acids has highlighted key residue K309, proposed to co-ordinate ATP phosphate group binding, and the motif N290F291R292 co-ordinating binding of the adenine ring at P2X1 receptors. To further develop the P2X receptor model and investigate the role of N290F291R292 and K309, adjacent nonconserved amino acids and the contribution of the 20 amino acid linker to the second transmembrane domain, we have undertaken substituted cysteine scanning mutagenesis (SCAM) of the region S286–I329 of the human P2X1 receptor. Analysis of the effects of these mutants on ATP potency, regulation by methanethiosulphonate compounds and determination of the accessibility of residues in a MTSEA-biotinylation assay has allowed us to identify amino acids involved in ATP binding, channel properties and information on the secondary structure of this section of the extracellular domain. p]Cysteine substitution was well tolerated and functional responses were recorded from all of the mutant channels. R295 and R305 however produced small currents of 16 and 94 nA respectively (10mM ATP). Substitution by cysteine reduced ATP potency at N290C, F291C, R292C and K309C receptors (∼125, 135, 20 and 240 fold respectively) supporting previous alanine mutagenesis. At the majority of residues (27/44) MTSEA (+ve charge) had no affect on ATP activation but caused a large inhibition of ATP responses for mutants N290C, R292C, G321C and I328C (4%, 4%, 41% and 41% relative to control respectively). In contrast at mutants F291C and K309C addition of MTSEA potentiated ATP evoked responses by 164% and 168% relative to control respectively. MTSES (-ve charge) had no effect on ATP response at the majority of residues (34/44) but significantly inhibited ATP response at N290C, R292C, K309C, D316C and G321C (41%, 10%, 6%, 69% and 44% relative to control respectively). F291C and A323C were both potentiated by 129% and 138% relative to control respectively. At N290, F291, R292 and K309 modulation of ATP responses by MTS reagents were coupled to a change in ATP potency indicating that these residues are involved in ATP binding. In contrast for mutants D316C, G321C, A323C and I328C, located in the linker region near to the channel pore, MTS reagents had no effect on ATP potency indicating involvement with gating or ionic flow. MTSEA biotin labels accessible cysteine residues and indicates possible local environment secondary structure around N290F291R292, K309 and the 20 amino acid linker region. Western blotting with P2X1 Ab on MTSEA biotin treated samples immunoprecipitated with streptavidin agarose showed three areas of accessibility, firstly 286–294 which contains the ATP binding region (N2-90F291R292), secondly 300–303 which contains a glycosylation motif (NGT) and thirdly the linker region close to the channel pore (D327–I329C). Interestingly K309 was not accessible to MTSEA biotin possibly due to the conformation of the binding pocket. These results provide a strong body of evidence supporting ATP binding at N290, F291, R292 and K309 residues of the P2X1 receptor. Supported by the Wellcome Trust

Co-release of ATP, NPY and NA from peripheral sympathetic nerves, modulation by CGRP and bradykinin

J.P. Huidobro-Toro and M.V. Donoso. Centro Regulación Celular y Patología, Instituto MIFAB, Departamento Fisiología, Facultad Ciencias Biológicas, P. Universidad Cató lica de Chile, Casilla 114-D, Santiago, Chile. jphuid@bio.puc.cl p]To ascertain the extracellular role of purines in synaptic transmission and particularly adenosine 5′ triphosphate (ATP) as a sympathetic co-transmitter, we investigated whether ATP is co-released together with noradrenaline (NA) and neuropeptide Y (NPY) from sympathetic nerve terminals. To examine the role of peptides as modulators of sympathetic co-transmission, we investigated whether calcitonin gene related peptide (CGRP) or bradykinin (BK) modulated the co-release of the sympathetic co-transmitters ATP, NA, and NPY. We used the prostatic half of the rat vas deferens from adult Sprague Dawley rats, a preparation enriched in sympathetic nerve endings. The tissues were mounted in superfusion baths and maintained with Tyrode buffer (37°C) supplemented with 1 mM desipramine, gassed with 95% O2/5% CO2 and superfused at a flow of 2 ml/min. Neurotransmitter release was elicited by electrical field stimulation (16 Hz, 60 V, 1ms during 1 min). All 3 sympathetic co-trans-mittters were assayed in a same sample aliquot from the collecting fluid. The overflow of ATP, NA and NPY were analytically quantified from a same aliquot sample of the perfusion buffer. As controls, non-stimulated buffer samples were also analyzed. Neurochemicals were determined by HPLC procedures: electrochemical detection quantified NA, while ATP and metabolites were detected as fluorescent derivatives. RIA quantified NPY. Electrical nerve ending depolarization elicited the co-release of ATP and NA plus NPY. The time course of ATP and NA overflow was different; the release of NA was considerably delayed as compared to ATP. These differences in the time course of overflow might indicate differential rates of diffusion from the tissues. The net outflow of ATP was 4–10 times larger than the outflow of NA. The application of 10 nM CGRP, a concentration that reduced the twitching of the electrically evoked muscle contraction, significantly reduced the overflow of ATP, NA and ir-NPY released by electrical stimulation. 30 nM BK, which increased 3-fold the electrically evoked smooth muscle twitching, halved the overflow of ATP and the outflow of adenosine from 3421 ± 526 to 1403 ± 131 pmol (p<0.05, n=4). The total NA and ir-NPY overflow was also reduced by BK, as shown in the following Table:





ATP (pmol)

95.79 ± 23.54 (7)

45.56 ± 35.36 (3)

37.74 ± 10.65 (3)

NA (pmol)

46.92 ± 9.22 (7)

19.24 ± 8.40* (3)

11.75 ± 4.98* (3)

Ir-NPY (fmol)

32.05 ± 11.20 (4)

15.56 ± 4.43 (2)

12.08 ± 0.91 (2)

In conclusion, electrical nerve ending depolarization elicited the overflow of extracellular ATP and NA consonant with their proposed co-transmitter role in sympathetic neuroeffector junctions. The much larger release of ATP might be related to its affinity for excitatory postjunctional P2X receptors, highlighting its role in sympathetic cotransmission. CGRP is a presynaptic sympathetic modulator, while BK has a dual action on sympathetic neuroeffector junctions highlighting the modulator role of neuropeptides in sympathetic co-transmission. The intracellular signaling mechanisms triggered by CGRP and BK to control sympathetic co-transmission is under investigation. Funded by FONDAP 13980001 and MIFAB.

CP-532,903 provides cardioprotection in two different mouse models of ischemia/reperfusion injury via an A3 adenosine receptor-mediated mechanism p]Tina C. Wan, Zhi-Dong Ge, John A. Auchampach. p]Department of Pharmacology and Toxicology, Medical College of Wisconsin, WI, USA twan@mcw.edu p]CP-532,903 is a 3′-aminoadenosine-5′-uronamide derivative shown previously to be a highly selective agonist for the human and rabbit A3 adenosine receptor (AR). The goal of this study was to determine the selectivity of CP-532,903 (CP) for the mouse A3AR and to characterize its cardioprotective profile in an in vivo mouse model of infarction and an isolated mouse heart model of global ischemia/reperfusion injury. In vitro binding assays conducted with HEK 293 cells expressing recombinant mouse ARs and [125I]I-AB-MECA determined that CP binds to the mouse A3AR with high affinity (9.6 ± 2.5 nM) and ∼100-fold selectivity versus the A1AR. In cAMP assays, CP did not stimulate cAMP production in HEK 293 cells expressing either the recombinant mouse A2aAR or the A2bAR at a concentration of 10 µM, demonstrating poor potency of the compound for mouse A2ARs. Infarct size in control mice subjected to 30 min of coronary occlusion and 24 h of reperfusion was 57 T 2% of the area at risk. In mice pretreated with 30 or 100 µg/kg of CP, infarct size was significantly reduced to 43% and 39% of the area at risk, respectively. Administration of CP produced a short-lived reduction in mean arterial blood pressure and significantly increased plasma histamine levels, responses likely due to A3AR-mediated degranulation of mast cells. In the isolated mouse heart model, coronary infusion of 10, 30 or 100 nM CP for 10 minutes prior to 20 min of global ischemia and 45 min of reperfusion produced a concentration-dependent improvement in recovery of left ventricular developed pressure and +/-dP/dt (∼30% at 30 and 100 nM) compared to vehicle-treated hearts. In contrast to the in vivo studies, infusion of CP in isolated hearts produced no changes in hemodynamic parameters. The selectivity of CP in the two mouse models for the A3AR was confirmed by repeating the studies using A3AR gene ‘knock-out’ mice. We conclude that the A3AR agonist CP provides protection against ischemia/reperfusion injury in mouse hearts by activating the A3AR.

Critical Role of 5′ — Ectonucleotidase (CD73) in Cardiac Ischemic Preconditioning p]Tobias Eckle1, Almut Grenz2, Marion Faigle1, Thomas Weissmüller1, Linda Thompson3, Sean P. Colgan4, Hartmut Oßwald2 and Holger K. Eltzschig1, 4 p]1Clinic of Anesthesiology and Intensive Care Medicine, and 2Institute of Pharmacology, Tübingen University Hospital, 3Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center, and 4Center for Experimental Therapeutics and Reperfusion Injury, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115, USA 1t.eckle@arcor.de p]Acute myocardial infarction (AMI) is among the leading causes of morbidity and mortality in the western countries. Cardiac ischemic preconditioning (IP) has been described as the strongest form of in vivo protection during AMI. Some evidence suggests that increased nucleotide phosphohydrolysis via CD73 [conversion of adenosine monophosphate (AMP) to adenosine] and nucleoside signaling via adenosine receptors1 contributes to the infarct size limiting effects of IP2. However, previous studies are limited to pharmacological approaches or indirect lines of evidence. Therefore, we established a murine model of cardiac IP and studied cardioprotective effects using a genetic approach. p]Following cardiac IP, CD73 mRNA was induced in preconditioned myocardium (18,4 ± 5,7− fold, n=6, p<0,05). Immunhistochemistry and enzyme activity measurements confirmed robust induction of CD73 protein and function by IP in preconditioned hearts. Using our murine model of cardiac IP, infarct sizes were reduced following IP by 69%, from 42.1 % to 13,5 % in wildtype mice. Alpha-beta-methylene-ADP (APCP, a highly selective CD73 inhibitor) treatment completely abolished the infarct size limiting effects of cardiac IP. Cd73−/− mice had significantly bigger infarct sizes at baseline. Moreover, the infarct size limiting effects of IP were completely abolished in these animals. Adenosine treatment of cd73−/− mice resulted in a partial reconstitution of the cardioprotective effects of IP (IP-Ado in cd73−/−-: 54% ± 3,0 vs. IP+Ado in cd73−/− 39,3% ± 3,3 infarct size; p<0,05). p]Taken together, our data reveal for the first time direct in vivo evidence that CD73 is critical for the observed cardioprotective effects of IP. Further studies will need to define the contribution of individual adenosine receptors to cardioprotection by IP with the goal to identify novel therapeutic targets in the treatment of medical conditions associated with myocardial ischemia.

Cross regulation of adenosine and uridine nucleotides salvage synthesis in rat and human brain cytosol p]Piero L. Ipata1, Lodovico Lutzenberger2 and Catia Barsotti1. ipata@dfb.unipi.it p]1Department of Biology, Section of Biochemistry; 2Department of Neuroscience, Section of Neurosurgery; University of Pisa, Pisa, Italy. p]The de novo pathways for purine and pyrimidine biosynthesis are cross-regulated, i.e., accumulation of a purine nucleotide (ATP) activates pyrimidine synthesis and inhibits purine synthesis, and vice versa. In bacteria the molecular mechanism relies on the modulation of a single enzyme, aspartate transcarbamylase (1). This enzyme is activated by elevated ATP levels, which can be viewed as a signal of purine sufficiency, and inhibited by CTP and UTP, a signal of pyrimidine sufficiency. In mammals the modulated enzyme is carbamoyl phosphate synthetase (2). However, it is well established that several tissues and organs, including brain, rely more heavily on the salvage synthesis of nucleotides from preformed bases and nucleosides, rather than on de novo synthesis from simple precursors (3). This raises the following question: how do these districts maintain the right balance between their purine and pyrimidine nucleotide pools? p]The rationale of our experimental procedure is the following. Since the purine and pyrimidine salvage processes occur at the base and nucleoside level, respectively (4, 5), rat brain cytosol is incubated in the presence of either radioactive adenine and cold uridine, to follow the rate of purine salvage, or with cold adenine and radioactive uridine, to follow the rate of pyrimidine salvage. We emphasize that this experimental procedure allows to measure the rate of purine salvage when also pyrimidine salvage is operative, and vice versa. The time course of radioactive purine and pyrimidine nucleotides formation is followed, in the presence of different concentrations of effectors. p]Our results show that in rat and human brain cytosol the two processes of purine and pyrimidine nucleotide salvage respond in an opposite manner to the fluctuation of intracellular [purine nucleoside triphosphates]/[pyrimidine nucleoside triphosphates] ratio (PUR/PYR ratio). The metabolic sensor is a two enzyme system which maintains uridine homeostasis, composed of uridine phosphorylase (UPase) and uridine kinase (Uk). When the activity of Uk is inhibited by relatively high UTP and CTP (6) concentration (low PUR/PYR ratio), uridine accumulates and the equilibrium of the UPase reaction is shifted towards uridine phosphorolysis. The ribose-1-P produced forms PRPP, an obligate substrate of adenine salvage (7). At a high ratio these effects are reversed. Uridine kinase becomes fully active, and pyrimidine salvage is favoured over purine salvage.

Cross-talk between human cysteinyl-leukotriene and extracellular nucleotide receptors: implication in inflammation

Valérie Capra1, Maria Rosa Accomazzo1, Simona Citro1, Liaman K. Mamedova3, Kenneth A. Jacobson3, Maria Pia Abbracchio2 and G. Enrico Rovati1

1Laboratory of Molecular Pharmacology, Section of Eicosanoids Pharmacology and 2Laboratory of Molecular and Cellular Pharmacology of Purinergic Transmission, Department of Pharmacological Sciences, University of Milan, Via Balzaretti, 9, 20133 Milan, Italy 3Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892, USA. GEnrico.Rovati@unimi.it p]Cysteinyl-leukotrienes (cysteinyl-LTs) and extracellular nucleotides mediate inflammatory responses via specific G protein-coupled receptors (GPCRs), the CysLTR and the P2YR, respectively. Evidence has begun to accumulate for the existence of a multifaceted cross-talk between the cysteinyl-LT and the nucleotide systems, considering that these mediators accumulate at sites of inflammation and inflammatory cells express both classes of receptors. For example, in rat microglia, activation of P2Y1R and CysLTR mediate co-release of both nucleotides and cysteinyl-LTs. Furthermore, we recently demonstrated that two CysLT1R selective antagonists marketed for asthma therapy functionally interact with P2YR signaling pathways by inhibiting nucleotide-induced calcium mobilization. Herein, we investigated the molecular basis of desensitization and trafficking of the human CysLT1R in response to LTD4 or nucleotides. Exposure of the CysLT1R endogenously expressed in monocyte-macrophage like U937 cells to agonist induced a rapid homologous desensitization followed by receptor internalization. Activation of P2YR with ATP or UDP induced CysLT1R heterologous desensitization. Conversely, LTD4-induced CysLT1R activation had no effect on P2YR responses. Furthermore, ATP/UDP-induced CysLT1R desensitization was unable to cause receptor internalization and was dependent upon PKC, at variance with homologous desensitization. Hence, CysLT1R desensitization and trafficking are differentially regulated by its cognate ligand or by extracellular nucleotides. This cross-talk may have a profound physiological implication in the regulation of responses at the sites of inflammation, and may represent just an example of a feed-back mechanism utilized by the cells to fine tune their responses. Furthermore, it is also possible that other inflammatory stimuli heterologously desensitize the CysLT1R in a similar way, as is the case for the chemoattractant receptor system. p]Partially supported by FIRB2001 to G.E.R. and by the Italian CARIPLO Foundation (2004 Projects “Promuovere la ricerca scientifica e tecnologica in tema di salute e scienze della vita” to G.E.R. and M.P.A., Project coordinated by prof. Francesco Clementi, University of Milan, Italy).

CVT-3619, a novel partial A1 adenosine receptor agonist, lowers plasma free fatty acids, triglycerides and improves insulin sensitivity without cardiovascular effects. p]Arvinder K. Dhalla, Mei-Yee Wong, Melissa Santikul, Michelle Smith, John C. Shryock and Luiz Belardinelli. p]Department of Pharmacological Sciences, CV Therapeutics Inc. Palo Alto, CA. USA 94304. email:arvinder.dhalla@cvt.com p]Elevated plasma free fatty acid (FFA) levels have been linked to the onset of insulin resistance. A1 adenosine receptor (A1R) agonists are potent anti-lipolytic agents which inhibit adipose tissue lipolysis and lower circulating FFA levels. Many A1R agonists have been developed in the past for the purpose of lowering FFA; however, their concurrent effects on the cardiovascular system are major limitations for their therapeutic utility. In the present study we report that CVT-3619, a novel partial A1R agonist, causes reduction in circulating FFA levels and improves insulin sensitivity without any significant effect on heart rate and blood pressure. Male Sprague Dawley rats (250–300 gms) implanted with indwelling arterial and venous cannulas for blood sampling and drug administration were used in the study. CVT-3619 (1–10 mg/kg, PO dose) decreased both FFA and triglyceride (TG) levels (15–57%) in a dose-dependent manner. CVT-3619 did not have any significant effect on heart rate and blood pressure up to 10 mg/kg, PO. At 25 and 50 mg/kg dose of CVT-3619 there was a small (10%) but significant decrease in heart rate. The FFA lowering effects of three consecutive single doses CVT-3619 were not significantly different suggesting that there was no tachyphylaxis of the effect of CVT-3619. In addition, the anti-lipolytic effect of CVT-3619 was not associated with a rebound increase in FFA levels as seen with Nicotinic acid. TG secretion (mg/dl/min) by the liver was significantly (p≪0.001) reduced with CVT-3619 treatment (slope for control 5.5 ± 0.12; CVT-3619 3.7 ± 0.17). CVT-3619 also lowered FFA and TG levels in rats with impaired insulin sensitivity caused by 2 weeks of high fat diet feeding. An oral glucose tolerance test (OGTT) in high fat diet fed rats treated with CVT-3619 for 2 weeks showed improved insulin sensitivity as compared to untreated control rats as indicated by the area under the curve for the time-course of the changes in glucose and insulin plasma levels (AUCg · AUCi). In conclusion, CVT-3619 is a novel, partial, orally bioavailable A1R agonist that lowers circulating FFA and TG levels resulting in improved insulin sensitivity in a diet-induced model of insulin resistance with minimal cardiovascular effects.

Design and Synthesis of purine and pyrimidine nucleosides carrying phosphate mimics groups at the 5′-position as potential P2 receptor ligands1 p]Maria-Cruz Bonache, Lisa Buzzoni, Nunzia Ciliberti and Stefano Manfredini p]Department of Pharmaceutical Sciences, University of Ferrara, Via Fossato di Mortara 17–19, 44100 Ferrara, Italia p]Extracellular nucleotides regulate a wide variety of functional responses in many cell types by stimulation of both G-protein coupled receptors (P2Y) and ATP-gated ion channel receptors (P2X). Selective ligands for purinergic 166 Communications Springer and pyrimidinergic P2 receptors are urgently needed in order to investigate their physiological role and the pharmacological potential of such compounds2. A useful approach to the discovery of novel P2 receptor ligands may consists in the design of purine and pyrimidine nucleosides carrying phosphate mimics groups at the 50- position. These compounds should be featured with structural changes compatible for the recognicition by the receptors but stable to the enzymatic hydrolisys. Consequently, we have recently explored the possibility to obtain a new class of stable nucleotide analogs endowed with isosteric substitution of the diphosphate group. The substitution of the diphosphate moiety of natural substrates with phosphonoacetic acid ester and amide moieties should give a stronger interaction with the active site, thereby resulting in potent ligands. On these premises, on the hand, we have prepared new nucleosides analogs designed in order to obtain bioisosters at the diphosphate moiety. Taking into account our previous results have explored 5′-phosphono acetic acid, amide and ester analogs of inosine, adenosine, uridine, cytidine and ribofuranosiltimine. In this communication, the synthesis of these compounds will be reported.

Design and synthesis of selective A2b adenosine receptor antagonists: New mono-N-1 alkyl 8-(pyrazol-4-yl) xanthines p]Rao Kalla,a Elfatih Elzein,a Thao Perry,a Xiaofen Li,a Tennig Maa,b Arthur Gimbel,b Dewan Zeng,b and Jeff Zablockia p]a Department of Bioorganic Chemistry, b Department of Drug Research and Pharmacological Sciences, CV Therapeutics Inc., 3172 Porter Drive, Palo Alto, CA 94304, USA rao.kalla@cvt.com p]Adenosine is an endogenous ligand that binds to four adenosine receptor subtypes — A1, A2a, A2b, and A3. Activation of the A2b adenosine receprot (AdoR) on mast cells may play a putative role in asthma through mast cell degranulation leading to the release of inflammatory cytokines (e.g. interleukin-8). Furthermore, activation of A2b AdoRs on bronchial smooth muscle cells (BSMC) has been demonstrated to lead to the production of interleukin-6 and monocytic chemotactic peptide-1, a process that can be blocked by selective A2b AdoR antagonists. Therefore, potent and selective A2b adenosine receptor (AdoR) antagonists may play a beneficial role in understanding the role of A2b AdoRs in physiological and pathological conditions and as a potential treatment for asthma. In our efforts to identify a slective high affinity A2bAdoR antagonist, we explored 8-pyrazol-4-yl xanthine derivatives that represents a new class adenosine receptor antagonists.1 Previously, we reported the SAR of 1,3-disubstituted 8-pyrazolyl xanthine derivatives that displayed high affinity for the A2bAdoR and good selectivity.2,3 We have shown that substitution of the pyrazole nitrogen with a meta-substituted benzyl group (1) or with oxadiazoles (2) and isoxazoles increased the A2b AdoR selectivity. Herein, we describe our efforts to increase the A2b AdoR selectivity of 8-pyrazolyl xanthines by exploring the effects of mono-substitution at the N-1 and N-3 positions of the xanthine. The N-1 substituted 8-pyrazolyl xanthines 3 and 4 have displayed higher selectivity compared to their corresponding disubstituted derivatives 1 and 2. The synthesis of mono substituted 8-pyrazolyl xanthines and their SAR will be discussed in detail. Open image in new window

Determinants for The Camp Binding Site at The S-Adenosylhomocysteine-(SAH)-Hydrolase p]D. Kloor1, J. Kirschler1, H. Kalbacher2, S. Stevanovic3, M. Müller3, H. Osswald1 p]Depts. of 1Pharmacology and Toxicology, 2Biochemistry, 3Immunology, Eberhard-Karls Universität, 72074 Tübingen, Germany doris.kloor@uni-tuebingen.de p]SAH-hydrolase catalyzes the reversible hydrolysis of SAH to adenosine and homocysteine. One mol of the enzyme contains 4 mol of NAD+ which is the essential cofactor of the enzyme1. As recently shown purified SAH-hydrolase exhibits two adenosine binding sites one with a high and one with a low affinity. These adenosine binding sites are controlled by the enzyme-bound NAD+/NADH ratio. The existance of the two sepatrate binding sites were identified after covalent labeling with 8-azido-adenosine2. Since adenosine competes with cAMP at the high affinity binding site of the enzyme3, we examined in the present study the effect of cAMP on the enzyme activity and its binding characteristics to SAH-hydrolase in its native, NAD+ and NADH form. p]Methods: SAH-hydrolase was purified to homogeneity from bovine kidney with chromatographical methods. The activity of SAH-hydrolase was assayed in the direction of hydrolysis. With the spectrophotometric titration assay we determined the ability of adenosine and cAMP to reduce the enzyme bound NAD+ at 327 nm. To gain further insight into the role of NAD+ in cAMP binding we removed NAD+ of the enzyme by dialysis and the apo-enzyme was reconstituted either with 100% NAD+ or with 100% NADH. For saturation binding experiments the enzyme (10 2g/ml) was incubated with 3H-cAMP (0.01–5 2mol/l) two hours at room temperature in Tris/Hepes pH 7.4. To identify the cAMP binding site, SAH-hydrolase was incubated with [2-3H]-8-azido-cAMP. After irradiation the reaction mixture was digested by Asp-N or trypsin and the peptides purified by HLPC were sequenced.

Results: cAMP enhanced the hydrolytic activity of SAH-hydrolase by 30% at 5 µmol/l. In contrast to adenosine, cAMP did not reduces the tightly bound NAD+ in the titration assay. In the presence of cAMP the conversion of NAD+ to NADH by adenosine is decreased in a concentration dependent manner. Saturation experiments performed by increasing concentrations of 3H-cAMP and with the three differnt forms of SAH-hydrolase, native, NAD+ and NADH, indicated only one binding site with a high affinity. This binding site was identified after irradiation of the enzyme with [2-3H]-8-azido-cAMP. One photolabeled peptide was isolated as Trp310-Val325 from the native SAH-hydrolase and one peptide as Asn314-Gln324 from both, NAD+ and NADH forms. p]Conclusion: Our data show that the cAMP binding site of the SAH-hydrolase is independent of the NAD+/NADH ratio of the enzyme. We conclude that the cAMP binding site is identical with the high affinity binding site of adenosine. This implies that the inhibitory effect of adenosine on the SAH-hydrolase can be antagonized by cAMP. Whether this cAMP action may have physiological implications remains to be determined.

Determination of Glial Cell Activation and C-Fos Immunoreactivity in The Brain of Huntington Transgenic Mice p]Sara Cipriani, Alessia Melani, Marco Gianfriddo1 and Felicita Pedata p]Dept. Of Pharmacology, University of Florece, Italy; 1Sienabiotech S.p.A, Siena, Italy. scipriani@unifi.it p]Huntington disease (HD) is an autosomal, dominantly inherited neurodegenerative disorder characterized by progressive motor and cognitive disturbances caused by an expansion in CAG repeats in the IT15 gene which encodes the huntingtin protein. A pathogenetic role for excitotoxic cell death mediated by increased glutamatergic excitoxicity in the striatum has been proposed. Drugs able to modulate striatal levels of glutamate are thus candidates to protect striatal neurons from neurodegeneration. We have recently demonstrated that the extracellular concentration of adenosine increases in the striatum of HD transgenic (R6/2) mice and that the selective antagonist of adenosine A2a receptors, SCH 58261, directly administered in the striatum, significantly reduces glutamate outflow1. Furthermore we demonstrated that the p38 mitogen-activated protein kinase (MAPK) pathway is activated in the neurons of the striatum of HD transgenic mice1. The p38 MAPK is known to be a factor activated and inducing the expression of pro- and inflammatory mediators and involved in the early gene c-Fos phosphorylation. Products of Fos family are players in inducing inflammatory gene expression in glial cells. In HD patient brain microglial activation in the striatum and in the cortex2 and reactive astrogliosis in the striatum during the late stage of pathology3 were reported. p]A first aim was therefore to investigate glial cells and c-Fos activation in heterozygous transgenic R6/2 male and wild-type mice at different ages, 10–11 (presymptomatic phase) and 14 week old (symptomatic phase), and to evaluate the effect of selective adenosine A2a receptor antagonist SCH 58261. SCH 58261 was subchronically administered (0.01 mg/kg i.p.) at time: −20, −17 and −2 h from sacrificed. Mice were transcardically perfused with paraformaldheyde and brains were cut by cryostat into 30 2m thick slices. Microglial cells stained by isolectine-B4 and astrocytes immunostained by GFAP were not activated in the striatum of 10–11 (n=5) and 14 week old R6/2 mice (n=3). Astroglial cells were detected only in the cingolate cortex of 14 week old mice (n=3). In the brain of 10–11 week old R6/2 treated mice (n=5) specific c-Fos immunostaining was not changed in comparison to wildtype mice (n=6). On the other hand, specific c-Fos immunostaining was increased in the piriform cortex, but not in the striatum, of 14 week old mice (n=3) in comparison to wild-type mice (n=6). The selective antagonist of adenosine A2a receptors decreased c-Fos immunoreactivity in the piriform cortex of 14 week old mice (n=3). p]The results demonstrate that in transgenic R6/2 mice in the terminal phase, there is a modest activation of glial cells and activation of c-Fos early gene in the cortex, and that adenosine A2a antagonism reduces c-Fos activation in the cortex. (Grant by Italian Ministry of Health and Fondazione Monte dei Paschi di Siena, Italy).

Developing an assay to probe the activity of the P2X1 receptor with radio-labelled 2-azido ATP p]*Kelvin C. Agboh, Jonathan Roberts, Andrew Powell1 & Richard J. Evans p]Department of Cell Physiology & Pharmacology, University of Leicester, Leicester, LE1 9HN, UK; 1Gene Expression & Protein Biochemistry, GlaxoSmithKline, Stevenage, SG1 2NY, UK *kca2@le.ac.uk p]Mutagenesis of conserved residues at the P2X1 receptor has revealed reductions in the potency of ATP which could be attributed to either binding or gating effects. To discriminate between them, we developed an assay to probe binding at the P2X1 receptor using 2-azido ATP. Whole cell patch clamp recordings show that 2-azido ATP is a potent agonist at the P2X1 receptor. The ATP analogue contains a photo-active azido (−N3) group that is chemically inert until activation by UV light. Upon irradiation, a covalent linkage between the compound and the active site of the receptor is formed. Cross-linking should cause permanent P2X1 receptor occupation, therefore subsequent ATP applications should produce responses that are greatly reduced from the control. p]HEK293 cells stably expressing the P2X1 receptor were incubated in 30µM 2-azido ATP and UV irradiated for 3 min, with the UV source positioned 7.5cm above the cells. Under these conditions and following a 10 min wash, UV or ATP and UV had no effect on the amplitude of the P2X1 receptor response. However, UV irradiation with 2-azido ATP caused an 86% reduction in currents elicited by 1µM ATP application (EC50 concentration). Control responses were reduced from 3.47 ± 0.28nA (n =15) to 0.49 ± 0.09nA (n = 30). This reduction was maintained throughout the recorded 45min time period. The effects of cross-linking were prevented by pre-treatment with 300µM ATP or 100µM suramin, this provided 29% and 73% receptor protection respectively. p]To detect the binding of 2-azido ATP to the receptor, we used a 32P labelled version of the compound. HEK293 cells stably expressing the P2X1 receptor were treated with radio-labelled 2-azido ATP. To determine the level of non-specific binding, non-transfected HEK293 cells were treated similarly. After lysis, the cross-linked P2X1 receptor was isolated by immunoprecipitation with the P2X1 antibody (Alomone) and the samples were run on a 10% SDS gel. Exposure of the dried gel to autoradiography film directly provided evidence of 2-azido ATP binding at the P2X1 receptor, no binding to non-transfected HEK293 cells and the reduced binding which occurred at cells protected by excess ATP. Previously shown mutations of the NFR region (290–292) and the positive lysine at the 309 position to either an alanine or cysteine residue revealed a significantly reduced ATP potency when compared to the wildtype. When expressed in Xenopus oocytes, binding of the radio-labelled 2-azido ATP was significantly decreased in these mutants when compared to the wildtype. This decrease is indicative of a reduction in direct binding to the P2X1 receptor and provides further evidence that the NFR region and the lysine residue at 309 contribute to the binding site of the P2X1 receptor.

Development and structure-activity relationships of ectonucleotidase inhibitors p]A. Brunschweigera, J. Iqbala, M. N. Munkondab, J. Sévignyb, A. F. Knowlesc, and C. E. Müllera p]aPharmaceutical Sciences Bonn (PSB), Pharmaceutical Chemistry, Institute of Pharmacy, University of Bonn, Kreuzbergweg 26, Bonn, Germany bCentre de Recherche en Rhumatologie et Immunologie, Sainte-Foy, Québec, Canada cDepartment of Chemistry, San Diego State University, San Diego, CA 92182-1030, USA andreas.brunschweiger@uni-bonn.de p]Full abstract can only be sent late

Development of a novel, automated method for the characterization and screening of NTPDase inhibitors by in-capillary enzymatic microreaction p]Jamshed Iqbal1, Herbert Zimmermann2 and Christa E. Müller1 p]1Pharmazeutisches Institut Poppelsdorf, Universität Bonn, Kreuzbergweg 26, D-53115 Bonn; 2AK Neurochemie, Biozentrum der J.W. Goethe-Universität, Marie-Curie-Str. 9, D-60439 Frankfurt am Main, Germany. jamshed.iqbal@uni-bonn.de p]The ecto-nucleoside triphosphate diphosphohydrolases (EC represent a major and ubiquitous family of ecto-nucleotidases. They catalyze the sequential hydrolysis of the γ- and β-phosphate residues of nucleoside triand diphosphates, producing the corresponding nucleoside monophosphate derivatives.1 The activation of P2 receptors is controlled by ecto-nucleotidases capable of hydrolyzing nucleoside tri- and diphosphates.1 Inhibition of ecto-nucleotidases can thus result in a potentiation of purinergic signaling, supporting the notion that endogenous ecto-nucleotidases reduce the effective concentration of the released nucleotide.2 Inhibitors of ecto-nucleotidases eotidases could thus represent valuable tools for amplifying the biological effects induced by extracellularly released nucleotides.

Several methods have been used for the determination of Michaelis-Menten constants (Km values), and inhibition constants (Ki values for enzyme inhibitors) of NTPDases, including radioisotopic, HPLC and spectrophotometric assays. All of these methods are time-consuming and suffer from further serious drawbacks. Capillary electrophoresis (CE) has recently emerged as a versatile technique for the quantitative analysis of nucleotides3 and for the monitoring enzymatic reactions.{4 CE systems have been successfully applied for assaying enzyme activity, including the determination of Michaelis-Menten constants (Km values), and inhibition constants (Ki values for enzyme inhibitors), exhibiting a number of advantages over conventional methods. These include rapid separation of substrate and product, ultra-low sample volume requirements, and high throughput by automation. CE is particularly useful for investigating enzymatic reactions involving charged substrates or products, e.g. for the monitoring of phosphorylation or dephosphorylation reactions.

In the present study we have developed an easy, rapid, nonradioactive at-capillary-inlet CE method for the screening and characterization of NTPDases inhibitors.5 The enzymes used in this study were expressed in CHO cells. Different standard compounds were tested as NTPDase inhibitors by this new method and results were in close agreement with data obtained from standard methods. The newly developed CE-based NTPDases inhibition and characterization assay is suitable for automation using 96-well plates and can be used for the high throughput screening of novel potential NTPDase inhibitors and substrates.

Diadenosine polyphosphate analogue controls postsynaptic excitation in CA3- CA1 synapses via a nitric oxide (NO)-dependent mechanism

Sergei Melnik,1 Michael Wright,2 Julian A. Tanner,2 Timur Tsintsadze,1 Vera Tsintsadze,1Andrew D. Miller,2 and Natalia Lozovayaa1

1Department of Cellular Membranology, Bogomoletz Institute of Physiology, Ukraine; 2Imperial College Imperial College Genetic Therapies Centre, Imperial College London, UK. a.miller@imperial.ac.uk

Previously, we have described the modulatory effect of diadenosine polyphosphates Ap4A and Ap5A on synaptic transmission in the rat hippocampal slices, mediated by presynaptic receptors (Klishin et al., 1994). We have reexamined this phenomenon probing with a non-hydrolysable Ap4A analogue diadenosine-5′,5′-P1,P4-[β,β′-methylene] tetraphosphate (AppCH2ppA) of high purity and low salt content. In contrast to our previous report we now describe how AppCH2ppA at low micromolar concentrations exerts a pure postsynaptic effect giving strong non-desensitizing inhibition of orthodromically evoked field potentials (OFPs), without affecting the amplitude of excitatory postsynaptic currents (EPSCs) and antidromically evoked field potentials (AFPs) recorded in the CA1 zone of hippocampus. The effects of AppCH2ppA on OFPs are eliminated by a P2 receptor antagonist PPADS but not mimicked by purinoceptor agonists, α,β-methylene-ATP and ATP-γ-S, indicating that a P2-like receptor is involved but not one belonging to the conventional P2X/P2Y receptor classes. In contrast to AppCH2ppA, ATP inhibits the EPSCs. Diadenosine polyphosphate receptor (P4) antagonist, Ip4I, was unable to modulate AppCH2ppA effects. Thus, the PPADS-sensitive P2-like receptor for AppCH2ppA appears to selectively control dendritic excitation of the CA1 neurons. The specific nitric oxide (NO)-scavenger 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (PTIO) is shown to attenuate significantly AppCH2ppA mediated inhibitory effects, indicating that NO is involved in the cascade of events initiated by the AppCH2ppA. Further downstream mediation by adenosine A1 receptors is also demonstrated. Such spatially selective postsynaptic dendritic inhibition may influence dendritic electrogenesis in pyramidal neurons and consequently mediate control of neuronal network activity.

Diadenosine Polyphosphates Are Selective Vasoconstrictors In Human Coronary Artery Bypass Grafts.

AR Conant1, WC Dihmis1, AWM Simpson2

1The Cardiothoracic Centre, Liverpool NHS Trust, Thomas Drive, Liverpool, L14 3PE, UK 2Dept Human Anatomy and Cell Biology, University of Liverpool, L69 3GE UK awms@liv.ac.uk

Nucleotides released by degranulating platelets are potential causative agents of post-operative contraction in coronary artery bypass grafts. Diadenosine polyphosphates (ApnA) are released alongside mono-nucleotides and have activity at several P2 receptors. Diadenosine polyphosphates are relatively long-lived vasoactive molecules, as their stability is enhanced by the preferential metabolism of mono-nucleotides by ecto-nucleotidases. Therefore diadenosine polyphosphates may be the cause of local nucleotide-mediated vasoconstriction at sites of tissue injury and inflammation in coronary artery bypass grafts.

Following ethical committee approval, sections of human radial and internal mammary artery and saphenous vein were obtained surplus to surgery with informed patient consent. Changes in tension were recorded upon application of diadenosine polyphosphates or the P2X1 receptor agonist αβmethyleneATP.

Radial artery and saphenous vein demonstrated robust concentration-dependent vasoconstriction to Ap4A, Ap5A and Ap6A and αβmethyleneATP. Cross-desensitisation revealed that diadenosine-mediated vasoconstriction apparently occurred via the P2X1 receptor. In contrast, in the internal mammary artery αβmethyleneATP-mediated vasoconstriction was weak or absent in the majority of samples. Consistent with activity at the P2X1 receptor diadenosine polyphosphate-mediated vasoconstriction was also absent. However in radial artery and saphenous vein, responses to Ap5A could not be inhibited by PPADS but were inhibited by suramin or NF279 at high concentrations. Indeed in the presence of PPADS, Ap5A-mediated vasoconstriction was enhanced suggesting that ectonucleotidase activity was reducing the availability of Ap5A at the smooth muscle. Consistent with these findings enhanced Ap5A-mediated vasoconstriction occurred when Ap5A-responses were repeated in the presence of excess ATP.

Our data would implicate diadenosine polyphosphates as a potential cause of post-operative vasoconstriction in coronary artery bypass grafts. However this activity is limited to saphenous vein and radial artery grafts since in comparison the internal mammary artery is relatively insensitive to the diadenosine polyphosphates.

Diastereoselectivity of the P2Y11 receptor: molecular insights

Ecke, D.1, Nahum, V.2, Tulapurkar, M.E.1, Fischer, B.2, Reiser, G.1

1Institut für Neurobiochemie, Medizinische Fakultät, Otto-von-Guericke-Universität Magdeburg, Magdeburg, Germany

2Department of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel denise.ecke@medizin.uni-magdeburg.de

The P2Y1 and P2Y11 receptors are closest homologues among the family of G-protein-coupled P2 receptors and are activated by adenine nucleotides exclusively 1, 2. We stably expressed the P2Y11 receptor in 1321N1 cells and monitored the cells for the change in fluorescence intensity of the calcium indicator Fura-2 after agonist stimulation. By the use of this method we found a clear difference in potencies of several borano/sulfur-modified diastereoisomers of ATP. The P2Y11 receptor clearly prefers the (B) isomers (ATP-α-B Sp isomers and ATP-α-S Rp isomers) of the derivatives tested. The (B) isomers of all derivatives tested were more potent at the P2Y11 receptor than the corresponding (A) isomers (ATP-α-B Rp isomers and ATP-α-S Sp isomers) and the parent compounds. In former studies we have shown that ATP-α-S and ATP-α-B diastereoisomers display a stereoselective activity at the P2Y1 receptor 3, 4. Here we report that the P2Y11 receptor exhibits exactly the opposite chiral discrimination. Moreover, the (A) isomer of 2-MeS-ATP-α-B was found to be most selective at the P2Y1 receptor being 3000-fold more potent at this subtype than at the P2Y11 receptor. The 2-MeS-ATP-α-B diastere-oisomers are ineffective at the P2Y2 receptor5. Thus, the 2-MeS-ATP-α-B (A) isomer can be used for subtypeselective activation of the P2Y1 receptor in tissues expressing also the P2Y2 and P2Y11 receptors. The distinct opposite diastereoselective activity of ATP derivatives at the P2Y11 and the P2Y1 receptor will allow deciphering structural differences of the ligand recognition sites between these receptor subtypes. Presently, we are working on the molecular basis of the diastereoselectivity. By means of mutagenesis and rhodopsin-based homology modelling we aim to characterize the putative stereoselective nucleotide recognition site of the P2Y11 receptor. This will aid in the development of subtype-selective agonists.

Different behaviour of P2X7 Receptors in Cortical vs. Hippocampal Astrocytes

Bianco F., Colombo A, Mele R, Matteoli M. and Verderio C.

CNR-Institute of Neuroscience, Univ. of Milano Dept. of Medical Pharmacology, Center of Excellence on Neurodegenerative Diseases, Milano Italy. f.bianco@in.cnr.it

P2X7 is an ionotropic purinergic receptor with the characteristic to undergo transition from a selective channel to an aspecific pore, permeable to molecules up to 900 kD. Prolonged activation of the P2X7 receptor and pore transition can cause cell permeabilization and lysis.

The mechanisms leading to pore formation are just starting to be unravelled. One hypothesis suggests that pore formation derives from dilatation of the channel through recruitment of additional receptor subunits (Wiley et al, 1998; Virginio et al, 1999; Smart et al, 2002) while another hypothesis suggests that activation of P2X7 determines synthesis of a second messenger, which is responsible for pore opening. In this view, pore opening might occur through a membrane protein distinct from the purinergic receptor (Persechini et al., 1998; Gu et al, 2001).

In line with the recent observation that MAPK may be involved in pore formation. (Faria et al, 2005), here we show that the capability of P2X7 receptor to form an aspecific pore requires p38 MAPK activation, as pretreatment with the MAPK inhibitor SB, strongly prevented pore transition, as assayed by Yo-Pro uptake. We also provide evidence for different behaviour of the P2X7 receptor in primary astrocytes from two rat brain regions, cortex and hippocampus. While P2X7 receptor in cortical astrocytes undergoes transition from selective ionotropic channel to aspecific pore, P2X7 receptor in hippocampal astrocytes does not form a pore. The different ability to undergo pore transition affects cytokine release. Following pore formation in cortical astrocytes exposed to LPS, caspase-1 is activated, and the enzyme is able to process the pro-inflammatory cytokine IL1-beta which is released in the extracellular medium in its active form. P2X7 activation in hippocampal astrocytes does not result in IL1-beta secretion, although the pro-cytokine is present in the cytosol of astrocytes activated with LPS. The different mechanisms involved in P2X7 receptor activation, by affecting the ability of cortical and hippocampal astrocytes to release proinflammatory cytokines, might influence the onset on neuroinflammation during neurodegenerative processes in different brain areas.

Differential Effect of Cyclic Stress and Constant Stress on ATP Release in Human Airway Epithelial Cells

Brian Button and Richard C. Boucher

Cystic Fibrosis Research and Treatment Center, University of North Carolina, Chapel Hill, North Carolina, USA. Brian_Button@med.unc.edu

The airway surface liquid (ASL) layer lining the superficial cells of the airways is crucial for regulating mucociliary clearance and removal of noxious materials from the lungs. As a result of ion channel and fluid regulation, the height of the ASL is maintained at ∼7 µm, where cilia are fully erect, capable of proper cilia beating. Our overarching hypothesis is that the concentration of extracellular ATP and adenosine (ADO), via the activation of P2Y2 and A2b purinoceptors located in the luminal membrane, modulate the activity of ion channels involved in fluid secretion and absorption, to maintain the ASL at this level. The significance of ATP/ADO on ASL regulation in airway epithelia is underscored by experiments that demonstrate: 1) addition ATP/ADO induces ASL fluid secretion and 2) Removal of extracellular ATP/ADO induces unregulated ASL fluid absorption. We have previously demonstrated that mechanical stimulation of airway epithelial cultures, using cyclic stress to emulate the stresses experienced in the lung during normal tidal breathing, elicits ATP release, to levels that are sufficient to stimulate ASL secretion. In addition to cyclic stresses in the lung, constant (static) stress, such as those associated with bronchoconstriction of the airways, can also be observed. However, the ability of such constant stress to stimulate ATP release and fluid secretion, like cyclic stress, is unknown. The goal of this study was to directly compare ATP release and its subsequent effect on fluid secretion, on airway cultures under physiological levels of cyclic or static stress. In these studies, well-differentiated human airway cultures were exposed to 30 minutes of: 1) constant transepithelial pressure of 20 cmH2O, 2) cyclic pressure, oscillating between atmospheric pressure and 20 cmH2O at 20 cycles/min, or 3) atmospheric pressure only. To determine the rate of ATP release, a cocktail of ATPase inhibitors (ebselen, βγMeATP and AP5A) were added to the luminal solution in order to prevent ATP hydrolysis. The cumulative ATP present in the ASL at 30 minutes was assessed by microsampling the ASL and quantified using a conventional luciferin-luciferase assay. In control cultures, the ATP release rate was 378 ± 69 fmol/cm2/min, similar to previously reported values. In cultures undergoing cyclic stress, ATP release was significantly stimulated, to 2514 ± 823 fmol/cm2/min. In contrast, the rate of ATP release during constant stress was 422 ± 126 fmol/cm2/min, significantly less than cyclic stress at the same pressure (stress) magnitude. Further, exposures of airway cultures to static pressure (20 cmH2O for 24h) failed to induce a significant change in steady-state ASL height, as measured by confocal microscopy (7.5 ± 1.6 versus 7.8 ± 0.9 µm for control). However, cyclic stress (20 cmH2O / 20 cycles/min for 24h) resulted in a doubling of ASL height (to 14.2 ± 1.7 µm). The cyclic stress-induced secretion was completely inhibited by the pre-treatment with apyrase/ADA to inhibit accumulation of ATP/ADO (to 4.4 ± 0.2µm), demonstrating that ASL secretion requires the stimulation of ATP release, which was not observed in cultures under constant stress. These results provide the first evidence of differential effects of cyclic and static stress on ATP release from airway epithelial cells. While the mechanism by which airway epithelial cells differentiate between cyclic and constant stressremainsto be elucidated, this model system is useful in delivering the stimuli required to investigate this phenomenon. Supported by the Cystic Fibrosis Foundation.

Differential effects of A2A adenosine receptor gene transfer and cyclic AMP elevation on the induction of distinct NF-κB target genes in vascular endothelial cells

Elaine W. Strong and Timothy M. Palmer

Division of Biochemistry & Molecular Biology, IBLS, University of Glasgow, Scotland, U.K. 0311557s@student.gla.ac.uk

The A2A adenosine receptor (A2AAR) is a critical non-redundant suppressor of inflammatory responses in vivo. One important mechanism by which this seems to occur is suppression of the NF-κB pathway, which controls the induction of multiple genes encoding proteins that initiate and sustain pro-inflammatory responses. We have previously demonstrated that A2AAR gene transfer in human umbilical vein endothelial cells (HUVECs) suppresses NF-κB activation in response to stimuli such as TNFα and LPS by reducing the nuclear accumulation of p50/p65 heterodimers. To assess the functional sequelae of this phenomenon, we have compared the effects of A2AAR gene transfer in HUVECs on the accumulation of three important gene products known to be controlled at least in part by NF-κB; cyclooxygenase-2 (COX-2), E-selectin and vascular cell adhesion molecule-1 (VCAM-1). Characterisation of COX-2 induction in response to TNFα revealed that while sustained induction (24 hr) was solely dependent on activation of p38 MAP Kinase, earlier time-points at which induction was first detectable (8 hr) were also sensitive to inhibition of the NF-κB pathway. At this time-point, A2AAR gene transfer alone was sufficient to reduce TNFα-stimulated COX-2 induction compared with controls, while no effect of the A2AAR was detectable at 24 hr. Surprisingly, addition of the A2AAR-selective agonist CGS21680 to A2AAR-expressing HUVECs actually reversed the effect seen with receptor alone. Time-course experiments in control cells using elevated concentrations of the adenylyl cyclase activator forskolin demonstrated that cAMP elevation was sufficient to promote the transient induction of COX-2 even in the absence of TNFα. However, in the presence of TNFα, exposure to different concentrations of forskolin exerted a biphasic effect, suppressing COX-2 induction at low concentrations while enhancing it at higher concentrations. These data suggest that the effects of A2AAR expression with or without agonist on TNFα-mediated COX-2 induction can be accounted for completely by corresponding changes in cAMP. Comparative analysis with E-selectin revealed that induction of this gene was more straightforward, being sensitive to NF-κB inhibition alone with no contribution of any other signalling pathway being detectable. Moreover, while induction was suppressed by A2AAR gene transfer, the suppression was more pronounced in the presence of CGS21680 and correlated with an ability of cAMP-elevating stimuli to block TNFα-mediated E-selectin induction in control cells. Interestingly, TNFα-mediated induction of VCAM-1 was unaffected by either A2AAR gene transfer or pre-treatment of control cells with cAMP-elevating agents. Taken as whole, these data argue that the ability of the A2AAR to modulate the expression patterns of defined κB-sensitive target genes in HUVECs can be explained completely by the ability of the receptor to couple positively to adenylyl cyclase and elevate intracellular cAMP levels. They also suggest that the net effects of A2AAR gene transfer on specific κB-regulated target genes will vary depending on the nature of their control by cAMP elevation.

Differential Expression of E-NTPDases in Glioma Cell Lines in Relation to Astrocytes

M. R. Wink1, E. Braganhol1, A. S. K. Tamajusuku1, E.A. Casali1, G. Lenz2, L. F. Zerbini3, T. A. Libermann3, J. Sévigny4, A. M. O. Battastini1 and S. C. Robson5.

1Departamento de Bioquímica, ICBS, UFRGS, Rua Ramiro Barcelos, 2600-anexo, CEP 90035-003, Porto Alegre, RS, Brazil, 2Departamento de Biofísica, IB, UFRGS, Porto Alegre, RS, Brazil; 3BIDMC Genomics Center and New England Baptist Bone and Joint Institute, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA 02215, USA; 4Centre de Recherche en Rhumatologie et Immunologie, Centre Hospitalier Universitaire de Québec, Université Laval, Québec, Canada and 5Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, 99 Brookline Avenue, Boston, MA 02215,USA abattastini@terra.com.br

Extracellular nucleotides and nucleosides modulate glial cell growth and are important factors involved in inflammation, stroke and progression of tumor cells. It is known that ATP and adenosine induce proliferation in human astrocytoma cells and that ectonucleotidases play a key role in purinergic signaling regulation. We undertook a comparative study of the degradation of extracellular nucleotides by primary astrocytes cultures and glioma cell lines. We have previously observed that extracellular ATP and ADP degradation by glioma cell is lower than by astrocyte cultures. In the present work, we have analyzed the pattern of phosphohydrolysis of extracellular ATP by HPLC and have characterized the nature of E-NTPDases expression by RT-PCR, quantitative real-time PCR, immunocytochemistry as well as Western blotting analysis in astrocytes and glioma cell cultures. We noted that astrocytes generate AMP as 50% of the degradation product at 3 h, whereas this nucleotide did not represent more than 5% of extracellular product in the glioma cell lines. Among the nuclosides and the non-phosphorylated degradation products, normal astrocyte cultures generate twice as much inosine and hypoxanthine (14 and 15%) when compared to adenosine (7%). Glioma cell lines had almost undetectable levels of adenosine, generating inosine as the major non-phosphorylated degradation product. By quantitative real-time PCR, we identified Entpd2 (CD39L1) as the dominant Entpd gene expressed by rat hippocampal, cortical and cerebelar astrocytes, while C6 glioma cells exhibit low expression of all NTP-Dases investigated. The substantial decreases in ATP and ADP hydrolysis observed in glioma cultures were substantiated by the absence of NTPDase2. ATP is recognized as a mitogenic factor and potentiates the proliferation in human glioma cells. Therefore, we suggest that reduction of the ATP hydrolysis in glioma cell lines may be part of a process to grant high progression rates of these tumor cells. We propose that alterations in the ecto-nucleotidase expression may represent an important mechanism associated with malignant transformation of glioma cells.

Support: NIH grants HL-63972 (Dr. S.C. Robson) and CNPq, FAPERGS (Dr. A.M.O. Battastini)

Differential regulation of NTPdase1 and NTPdase2 by cholesterol.

A.Papanikolaou1, A.Papafotika1,2, A.Xilouri3, J.Mitsios4, A.Tselepis4, and S.Christoforidis1,2

Laboratory of Biological Chemistry, Medical School, University of Ioannina Biomedical Research Institute/Foundation for Research and Technology, Ioannina Department of Obstetrics and Gynecology, University Hospital of Ioannina, Ioannina Laboratory of Biochemistry, Department of Chemistry, University of Ioannina, Greece. me00897@cc.uoi.gr

Nucleoside triphosphate diphosphohydrolases (NTPDases) are a family of ectonucleotidases that differentially hydrolyze γ and β phosphate groups of extracellular nucleotides. NTPDase1 is the major ecto-nucleotidase at the luminal surface of blood vessels and hydrolyzes both triphospho- and diphosphonucleosides thereby terminating platelet aggregation in response to ADP. In contrast, NTPDase2 is a preferential nucleoside triphosphatase in the adventitia of blood vessels. By converting ATP to ADP, NTPDase2 activates platelet aggregation in the case of vessel damage.

The enzymatic activity of NTPDase1 depends on cholesterol-rich domains of the membrane which points to a NTPdase1-dependent protective role of cholesterol in thrombosis. If this is true, given the platelet-activating action of NTPDase2, cholesterol should exert an opposite effect on this NTPDase. To address this question we increased the cholesterol content of isolated heart membranes using cholesterol complexed with methyl-β-cyclodextrin and measured its effect on ecto-ATPase enzymatic activity. Interestingly, treatment of membranes with increasing concentrations of cholesterol resulted in a corresponding inhibition of the enzymatic activity of NTPdase2.

To obtain further insights into this inhibition, we measured the endogenous cholesterol levels in the heart membranes. We were surprised by the finding that the cholesterol content in heart membranes (where NTPdase2 is enriched) is 10 times lower than the cholesterol amount in placental membranes (where NTPdase1 is enriched). It is therefore likely that NTPDase2 is localized in cholesterol-poor domains for maintaining a high enzymatic activity.

A key function of NTPDase2 is to promote platelet plug formation in the case of vessel injury. To examine the role of cholesterol on NTPDase2-dependent activation of platelet aggregation, we performed platelet aggregation assays in the presence of heart membranes incubated with cholesterol. High levels of platelet aggregation were obtained when heart membranes were added to platelets and ATP, which were significantly reversed when heart membranes had been incubated with cholesterol.

The above data raise the possibility that in pathological cases associated with high cholesterol levels, such as hypercholesterolemia and atherosclerosis, a cholesterol-dependent protective effect against thrombosis is of crucial importance. Consistent with this view, we found that, similarly to cholesterol, oxysterols and oxidized LDL exert an inhibitory effect on NTPdase2. Therefore, a concomitant differential control of NTPdase1 and NTPdase2 activities by cholesterol may result in a fine tuning of extracellular nucleotide levels, providing a tight control of purinergic receptor signaling. Such a regulatory mechanism is expected to apply not only in thromboregulation, but also in other functions regulated by purinergic receptors where the fine balance of ATP/ADP ratio is of major importance.

Discovery of new potent, non-nucleotide ligands at human P2Y11-receptors results in NF546, the first non-nucleotide P2Y receptor agonist, and NF340, the so far most potent and selective P2Y11 antagonist

Sabine Meis, Darunee Hongwiset, and Matthias U. Kassack

Institute of Pharmacy, University of Bonn, An der Immenburg 4, D-53121 Bonn (Germany) sabine.meis@uni-bonn.de; kassack@uni-bonn.de

P2Y11 receptors are G protein-coupled receptors coupled to both, the phosphoinositide and the cAMP pathway 1. Up to now, only suramin as a non-selective and low potency compound (pKi = 6.52) and NF157 recently introduced by our group as a more selective and potent ligand (pKi = 7.35) have been described as P2Y11 antagonists 2. Non-nucleotide agonists at P2Y receptors are not yet available. The aim of this study was to further improve the potency and selectivity of NF157 and to search for non-nucleotide P2Y11 agonists. A large number of sulfonic and phosphonic acid derivatives were screened at P2Y11 receptors recombinantly expressed in 1321N1 astrocytoma cells. Compounds were first studied with a previously described microplate-reader based Ca2+-assay 3. Compounds active in the Ca2+-assay were then also examined in a cAMP luciferase reporter-gene assay at P2Y11 receptors and further tested for selectivity at other P2Y-receptors.

The tetrasulfonic acid derivative NF340 was identified as a competitive antagonist with nanomolar potency (pKi = 7.77) which is 2.6fold more potent than NF157. Concentrations up to 100 µm showed no effects at other P2Y receptors (P2Y1, P2Y2, and P2Y6, recombinantly expressed in 1321N1 astrocytoma cells). Schild analysis with ATPγS as agonist gave a slope not significantly different from unity. The pA2 value was estimated as 8.07. The screening also identified the first non-nucleotide agonist at P2Y receptors. The tetraphosphonic acid derivative NF546 showed almost the same intrinsic activity (95%) at P2Y11 receptors as the standard agonist ATPγS in both assays, Ca2+ and cAMP. An EC50 of 542 nM was obtained in the Ca2+-assay. Schild analyses with NF546 as agonist and NF157 or NF340 as antagonists were performed. The slopes of both Schild plots were not significantly different from unity, and the pA2 values of NF157 and NF340 using NF546 as agonist showed no difference from the results using ATPγS as agonist: NF157 / NF546: pA2 = 7.66; NF157 / ATPγS: pA2 = 7.77; NF340 / NF546: pA2 = 8.00; NF340 / ATPγS: pA2 = 8.06. NF546 showed no effect at other P2Y receptors up to 10 µm.

In conclusion, NF340 was identified as a selective, non-nucleotide antagonist at P2Y11 receptors with improved potency over the recently described NF157. NF546 is the first non-nucleotide P2Y agonist and selective for P2Y11 receptors. Both compounds may serve as useful tools in studying the physiological role of P2Y11 receptors.

This project was supported by Bischoefliche Studienfoerderung Cusanuswerk, Deutsche Forschungsgemeinschaft (GRK677), and Deutscher Akademischer Austauschdienst.

Discovery of Novel P2X7 Antagonists

G. Steven Dodson1, John Kincaid2, Yeyu Cao2, Candace Chi3, Yenfung Fang2, David Hackos4, Sami Hussain4 Carl Kaub2, Luna Lui3, Eileen Rose1, Jennifer Shumilla5 Qingling Zhang3, Michael G. Kelly.

1Department of Discovery Biology, 2Department of Medicinal Chemistry, 3Department of Drug Metabolism and Pharmacokinetics, 4Department of in vitro Pharmacology, 5Department of in vivo Pharmacology Renovis, Inc. dodson@renovis.com, kincaid@renovis.com

The P2X7 receptor is a ligand-gated ion channel that mediates a non-selective cation conductance in response to extracellular ATP. Extended exposure to agonist leads to the formation of a pore permeable to large molecular weight dyes. P2X7 is expressed in a fairly limited set of cell types, primarily the hematopoietic lineages and glia, where it is thought to play a role in inflammatory process important in human disease states. In monocytes, macrophages and microglia primed with an inflammatory signal P2X7 functions to initiate the post translational processing and release of leaderless cytokines such as IL-1β. Published work in P2X7 knockout mice shows that the absence of P2X7 activity in vivo can modulate an inflammatory response and attenuate the severity of mAb induced arthritis. In addition, absence of P2X7 attenuates mechanical and thermal hyperalgesia in the Seltzer model of neuropathic pain and inflammatory hypersensitivity in FCA induced inflammation.

Our continued interest in the discovery and development of novel analgesic and anti-inflammatory therapeutics has led to the identification of multiple series of novel, selective and bioavailable P2X7 receptor antagonists. Herein, we will disclose the in vitro and in vivo characterization of RN-6189, an early biological lead that inhibits human P2X7 current in transfected HEK293 cells with an IC50 of 34 ± 1.6 nM. In addition, RN-0276189 inhibits both 2′,3′-O-(4-benzoylbenzoyl)-adenosine 5′-triphosphate (BzATP) dependent pore formation and IL-1β release in THP-1 cells in the same concentration range. Further property based drug design has led us to identify novel compounds with improved potency and pharmacokinetic properties that are currently in preclinical evaluation.

Distinct effects of ticlopidine and clopidogrel on vascular ecto-nucleotidases and impact on haemostasis.

Lecka J.1, Rana M.S.1, Zimmermann H.2 and Sévigny J.1

1Centre de recherche en Rhumatologie et Immunologie, Université Laval, Ste-Foy, Québec, Canada.

2Biocenter, J.W. Goethe-University, AK Neurochemistry, Frankfurt am Main, Germany. e-mail: joanna.lecka@crchul.ulaval.ca

Ticlopidine (Tyklid) and clopidogrel (Plavix) are antiaggregatory drugs that irreversibly inhibit P2Y12 receptor1. The real active metabolites that reduce platelet activation are produced by the liver after daily administration of the drugs2. P2Y12 receptor is activated by extracellular ADP which leads to platelet aggregation. Adenosine, the product of ADP dephosphorylation, inhibits platelet aggregation via activation of A2a receptors3. Blood level of both purine ligands is regulated by ectonucleotidases, mainly NTPDase1 and ecto-5′-nucleotidase4. As both of these enzymes have an important implication in haemostasis, we examined in vitro whether ticlopidine and clopidogrel could affect their biochemical activities. Concentrations of 30 µm ticlopidine and 10 µm clopidogrel have been reported in the plasma of treated patients. At this concentration, ticlopidine reduced ATPase and ADPase activity by 30% while 100 µm of both drugs inhibited 60% of ADP hydrolysis; they had no effect at 10 µm. Since NTPDase1 is an important inhibitor of platelet aggregation, inhibiting this enzyme would be expected to generate undesired effects in treated patients. Indeed, the antiagreggatory effect of human NTPDase1 was significantly weaker in the presence of ticlopidine, as seen in platelet aggregation assays. Additionally, ticlopidine (but not clopidogrel) inhibited rat ecto-5′-nucleotidase by 11% at 100 µm and 50% at 1 mM level. Interestingly, the human ortholog of ecto-5′-nucleotidase was insensitive to both drugs. We observed that the inhibition of the vascular ectonucleotidases by ticlopidine is direct and did not require preincubation. Our data suggest that the action of ticlopidine in haemostasis is complex and not exclusively antiaggregatory while its analog clopidogrel may be more suitable for therapeutic use.

Distribution and Regulation of the nucleoside transporter, ENT1

Kay Barnes1, Halina Dobrzynski2, Sophie Foppolo1, Paul R. Beal3, James Tellez2, William C. Claycomb7, Carol E. Cass4,5,6, James D. Young4, Rudi Billeter-Clark1, Mark R. Boyett2 and Stephen A. Baldwin1.

From the 1Institute of Membrane and Systems Biology, University of Leeds, Leeds, LS2 9JT, United Kingdom, the 2Division of Cardiovascular and Endocrine Sciences, University of Manchester, Manchester M13 9XX, United Kingdom, the 3Department of Biology, University of York, United Kingdom, the 4Membrane Protein Research Group, Departments of Physiology and 5Onconlogy, University of Alberta and the 6Cross Cancer Institute, Edmonton, Alberta T6G 2H7, Canada, 7Department of Biochemistry and Molecular Biology, Louisiana State, University Health Sciences Center, New Orleans, Louisiana 70112, U.S.A. k.barnes@leeds.ac.uk

ENT1 is a well-characterized member of the equilibrative (SLC29) nucleoside transporter family that has an important role in regulation of adenosine concentrations in the vicinity of extracellular purinergic receptors. ENT1 has been designated as belonging to the es (equilibrative inhibitor-sensitive) subtype of nucleoside transporters as it is potently inhibited by NBMPR (nitrobenzylmercaptopurine riboside) in the nanomolar range. It has 11 transmembrane helices with a large intracellular loop joining transmembrane domains 6 and 7, a cytoplasmic N-terminus and and extracellular C-terminus. Open image in new window

In this study we have used qualitative and quantitative immunological approaches to examine the relative distribution of rENT1 in the ventricles, atria and SA node of the rat heart. In parallel experiments using RT PCR and tissue samples prepared from the same rats hENT1mRNA transcripts showed a similar distribution to the transporter protein. Although little is known about the mechanism of nucleoside transporter regulation, activity of the transporter can be modulated by activators and inhibitors of protein kinase C and protein kinase A and by the serine/threonine protein kinase, casein kinase II. Using a bioinformatic approach we have identified potential phosphorylation sites in hENT1 that may be important in transporter regulation and expression constructs encoding hENT1 mutants lacking the potential phosphorylation sites have been prepared.

Sequential activation of adenosine A1 receptors terminate inhibition of acetylcholine release caused by ADP-sensitive P2Y1 receptors on myenteric motoneurons

Margarida Duarte-Araújo, M. Alexandrina Timóteo & Paulo Correia-de-Sá

Laboratório de Farmacologia, Unidade Multidisciplinar de Investigação Biomédica (UMIB), Instituto de Ciências Biomédicas de Abel Salazar (ICBAS) — Universidade do Porto, Portugal. mdcma@icbas.up.pt

Purine nucleotides and nucleosides are responsible for modulating gastrointestinal function. Besides the direct effect of ATP on neuronal P2X receptors, the nucleotide may indirectly control nerve excitability due to the formation of metabolites. In the myenteric plexus, ATP actions are rapidly terminated by intense ecto-NTPDase 1 (ecto-diphosphohydrolase) activity generating AMP, which is then converted into adenosine by ecto-5′-nucleotidase. Alternative formation of ADP by ecto-ATPase might be physiologically relevant to restrain acetylcholine (ACh) release from myenteric motoneurons via P2Y1 purinoceptors activation. The co-existence of inhibitory P2Y1 and adenosine A1 receptors in the rat myenteric plexus lead us to investigate their role to control [3H]-ACh release from stimulated (5 Hz, 200 pulses) cholinergic motoneurons.

The experiments were performed at 37°C on longitudinal muscle-myenteric plexus of rat ileum, superfused with gased (95% O2 + 5% CO2) Tyrode's solution. The release of [3H]-ACh from myenteric motoneurons loaded with [3H]-choline (2.5 µCi/mL), was induced by electrical field stimulation (5 Hz, 200 pulses, 0.5 ms, 60 V) (S1 and S2). Test drugs were added 15 minutes before S2, and their effects in transmitter release were expressed by the S2/S1 ratios. When we evaluated changes in the effect of tested drugs induced by a modifier, the latter compound was applied 15 min before starting sample collection and hence it was present during S1 and S2.

Selective blockade of A1- and P2Y1-receptors respectively with 1,3-dipropyl-8-cyclopentyl xanthine (DPCPX, 10 nM) and MRS2179 (300 nM) increased [3H]-ACh release by 43±12% (n=3) and 27±3% (n=4), respectively. A reduction of tritium outflow was observed by activating adenosine A1 receptors, with R-N6-phenylisopropyl adenosine (R-PIA, 300 nM, −36±4%, n=4), or P2Y1 purinoceptors, with ADPβS (30 µm, −27±3%, n=4). The inhibitory action of ADPβS (30 µm) was blocked by MRS2179 (300 nM), but the P2Y1-receptor antagonist was devoid of effect on R-PIA (300 nM) inhibition. On the contrary, DPCPX (10 nM) attenuated inhibition by R-PIA (300 nM), while significantly enhancing (to 42±5%, n=4) ADPβS (30 µm)-induced depression. While ADPβS (30 µm) failed to modify the action of R-PIA (300 nM), activation of adenosine A1 receptors occluded the inhibitory effect of ADPβS (30 µm).

We provided evidence showing that adenosine exerts a fine-tuning control of tonic ACh release inhibition mediated by released adenine nucleotides at the myenteric synapse. The results indicate that over stimulation of inhibitory P2Y1 purinoceptors is cut-short by sequential activation of inhibitory A1 receptors by endogenously generated adenosine.

Work supported by FCT (POCTI/45549/FCB/2002, participation of FEDER funding).

Early and transient alteration of adenosine A2A receptor signalling in a mouse model of huntington disease

Alessia Tarditi, Alessandra Camurri, Katia Varani1, Pier Andrea Borea1, Ben Woodman2, Gillian Bates2, Elena Cattaneo and Maria P. Abbracchio

Department of Pharmacological Sciences, University of Milan, Italy, and 1Department of Clinical and Experimental Medicine, Pharmacology Unit, University of Ferrara, Italy 2 Department of Medical and Molecular Genetics, GKT School of Medicine, King's College London, UK

Huntington Disease (HD) is a dominantly-inherited neurodegenerative disorder featuring progressive worsening chorea, psychiatric disturbances and cognitive impairment due to brain cell loss. A2A receptors expressed on GABAergic striatal neurons have been suggested to play a pathogenetic role. Previous data demonstrated the presence of an aberrant alteration of A2A receptor-dependent adenylyl cyclase in an in vitro model of the disease (striatal cells expressing mutant huntingtin) and in peripheral circulating cells of HD patients, suggesting that the aberrant A2A receptor phenotype may represent a novel potential biomarker of HD. Despite evidence suggesting a facilitatory role of A2A receptors in neuronal death, a direct correlation between the peripheral A2A receptor change in HD subjects and the status of this receptor in HD brain is still lacking. One of the best characterized animal models of HD is represented by the R6/2 mice. We thus investigated in this model the presence, expression and functionality of striatal A2A receptors at different developmental stages in comparison with age-matched wild type animals. A transient increase in A2A receptor density (Bmax) with no changes in affinity (KD) and A2A receptor-dependent cAMP production at early presymptomatic ages (7–14 postnatal days) was found. The increase of A2A receptor-mediated adenylyl cyclase responsiveness in R6/2 mice was even more marked, as demonstrated by an EC50 value of the A2A receptor agonist CGS 21680 ten fold lower with respect to corresponding EC50 value in wild type mice (175±23 nM, with respect to 1922±197 nM). Both alterations normalized to control values starting from postnatal day 21. In contrast, A2Areceptor mRNA, as detected by real time PCR, dramatically decreased starting from PND21 until late symptomatic stages (12 weeks of age). The discrepancy between A2A receptor expression and density suggests compensatory mechanisms. These data, reproducing ex vivo the previous observations in vitro, support the hypothesis that an alteration of A2A receptor signalling is present in HD and might represent an interesting target for neuroprotective therapies.

Sponsored by the Italian Ministry of Health (progetto “Ruolo del recettore A2A dell'adenosina e metabotropici del glutammato nelle demenze da neurodegenerazione striatale”, Alz4) and by the Hereditary Disease Foundation.

Ecto-5′nucleotidase (CD73) activity and expression in the rat ileum

Anna Bin1, Maria Cecilia Giron1, Sabrina Etteri1, Francesca Bongiovanni1, Chiara Florio2, Rosa Maria Gaion1

1Department of Pharmacology and Anesthesiology, University of Padova; 2Department of Biomedical Sciences, University of Trieste anna.bin@unipd.it


Ecto-5′nucleotidase (CD73) is a purine salvage enzyme located on the external surface of plasma membrane in many cell types. CD73 is responsible for adenosine formation from AMP and has a regulatory role in neurotransmitter release, functions to control blood flow and may play a role in antiinflammatory responses1, 2, 3. Concerning the intestinal tract, expression of CD73 has been demonstrated in human duodenal mucosa, intestinal epithelial cells and in the longitudinal muscle strips of guinea pig ileum1,4.


The purpose of the present study was to evaluate the metabolism of exogenous AMP by rat ileum strips (ISs) and the role of CD73 in the production of extracellular adenosine (ADO) in the intestine. Expression of CD73 in ileum smooth muscle cells (ISMCs) was also examined.


Rat ISs were incubated in vitro in 2 ml of Tyrode solution with exogenous AMP (50 µM) for 5 or 60 minutes and 1 ml of the medium was used for the determination of AMP, ADO, inosine and IMP by HPLC. ISMCs were prepared according to the method of Khan5 and the presence of CD73 was detected by confocal fluorescent microscopy using anti-CD73 antibody.


AMP added to the incubation medium was metabolised by 82% and 100% within 5 and 60 min, respectively. Treatment with α,β methylene-ADP (AOPCP, 200 µM), a CD73 inhibitor, allowed the recovery of endogenous AMP and IMP (at 5 min 0.2±0.08 and 0.8±0.2 µM, respectively, and at 60 min 0.5±0.1 and 2.2±0.3µM, respectively). AOPCP also reduced the clearance of exogenous AMP, whose concentrations at 5 and 60 min rose to 18.7±1.8 and 4.4±0.9 2M, respectively. The addition of AMP caused a marked increase in ADO concentration in the medium. This increase was inhibited by AOPCP, indicating that the nucleoside is rapidly formed from AMP. Inosine showed the same trend as ADO, in accordance with the high activity of intestinal ecto-ADA6. Addition of AMP in AOPCP-pretreated samples did not significantly modify IMP levels, showing that the contribution of AMP deaminase to exogenous AMP metabolism is irrelevant. Immunocytochemical analysis demonstrated the high expression of CD73 on the surface of ISMCs.


These results suggest that CD73 is a critical enzyme responsible for the generation of adenosine. Its expression on ISMCs, shown here for the first time, points out the possible key role in the control of intestinal motility. This is the first step in the development of further studies aimed to understand how CD73 expression can be affected by drugs and by pathological conditions.

Ecto-5′-nucleotidase (CD73) and Mucosal Inflammation

Sean P. Colgan1, Linda F. Thompson2, Andreas Robinson1 and Nancy A. Louis1,3

1Center for Experimental Therapeutics, Brigham and Women's Hospital and Harvard Medical School, Boston, MA USA; 2Immunobiology and Cancer Research Program Oklahoma Medical Research Foundation, Oklahoma City, OK; 3Division of Newborn Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA USA. colgan@zeus.bwh.harvard.edu

Sites of inflammation are characterized by significant shifts in metabolic activity1. Shifts in energy supply and demand can result in diminished delivery and/or availability of oxygen, leading to inflammation-associated tissue hypoxia and metabolic acidosis. These shifts in tissue metabolism, as indicated by previous studies, are frequently associated with vasculitis and profound recruitment of inflammatory cell types, particularly myeloid cells such as neutrophils (PMN) and monocytes. Under such conditions, we and others have observed activation of the global hypoxia regulatory transcription factor hypoxia-inducible factor (HIF). A significant HIF target gene during mucosal inflammatory disease (e.g. inflammatory bowel disease, colitis) is CD732. Previous studies have determined that CD73 is regulated by HIF at the gene promoter level3. Since tissue profiling studies revealed highest expression of CD73 in mucosal tissues, particularly the intestine4, we have addressed the role of CD73 in model inflammatory disease in vitro and in vivo. Studies using intestinal epithelial cells have revealed that high expression of CD73 is protective for a number of epithelial properties (e.g. barrier function, ion transport), likely through generation of adenosine at the epithelial apical surface. Inflammatory studies in mice conditionally lacking the alpha subunit of HIF-1 in intestinal epithelial cells have shown that the loss of functional HIF-1 correlates with decreased expression of CD73 and enhanced inflammatory disease in these mice. More recent studies have addressed the role of CD73 colitic disease progression. These ongoing studies have revealed that the induction of colitis increases epithelial expression of CD73 mRNA and activity during the acute phase colitic disease. Studies utilizing Cd73-null mice4 have revealed that relative that the loss of CD73 correlated with more severe clinical symptoms of colitis (weight loss, colon length), and increased levels of inflammatory markers (neutrophil numbers). Mechanisms of increased susceptibility with decreased CD73 will be discussed. Taken together, these studies provide unique insight into tissue microenvironmental changes during model inflammatory disease and identify CD73 as a critical control point during mucosal insult.

Ecto-ATP as a danger signal: Overexpression of P2RX7 on CD4+CD25+ regulatory T cells correlates with a higher susceptibility to ATP induced cell death

Sahil Adriouch, Sandra Hubert, Friedrich Haag, Olivier Boyer and Michel Seman

EA1556 University Denis Diderot-Paris7 and INSERM U519, Faculty of Medicine and Pharmacy, 22 Boulevard Gambetta, 76000, Rouen, France michel.seman@univ-rouen.fr

The P2X7 receptor is expressed on most cells of the immune system including T lymphocytes. On these cells, P2X7 expression can be modulated depending on cell activation. Allelic polymorphism of P2X7 such as the P451L natural mutation in the mouse also affects its expression on the surface of naive T cells. In both wild type and 451L mutant1, a higher density of P2RX7 is found on the subset of CD4+ CD25+ CD62Lhigh Foxp3+ regulatory T cells compared to naïve CD25 T cells. Overexpression of P2RX7 may thus represent another marker of Tregs in both mouse and man. Consistently, Tregs are more prone to apoptosis induced by exposure to ATP or the P2RX7 agonist bzATP. In the mouse, P2RX7 activation can also be induced by ART2 mediated ADP-ribosylation2. In C57Bl/6 mice which harbour the 451L mutant but have a high density of ART2 on their surface, ART2 substrate, NAD+, induces the death of CD4+ CD25+ T cells but not that of CD4+CD25 cells. By contrast, Tregs from BALB/c mice, which harbour the 451P wild type P2RX7 but a very low density of ART2, are resistant to NAD induced cell death but not to ATP induced cell death. Results suggest that ATP also plays a new role as a danger signal in the immune system. ATP released during inflammation and infection may contribute to the induction of the adaptative immune response by inducing the death of suppressive regulatory T cells.

Supported by grants from the Ministère de la Recherche, Ligue Nationale contre le cancer, Association pour la recherche sur le cancer.

Ecto-nucleotide pyrophosphatase/phosphodiesterase as part of a multiple system for nucleotide hydrolysis by platelets from rats: Kinetic characterization and biochemical properties

Cristina Ribas Fürstenau1, Danielle da Silva Trentin1, Maria Luiza, Morais Barreto-Chaves2 and João José Freitas Sarkis1

1Departamento de Bioquímica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil 2Departamento de Anatomia, Universidade de São Paulo, São Paulo, SP, Brazil jjsarkis@plugin-com.br

Platelets are known to play a major role in the maintenance of endothelial integrity and hemostasis1. ADP acts to potentiate platelet aggregation2. ATP has been considered as a competitive inhibitor or ADP-induced platelet aggregation3. The action of nucleotides is terminated by a surface-located enzyme cascade that sequentially degrades nucleoside 5′-di and 5′-triphosphates to their respective nucleosides4. Since nucleotides exert different responses in diverse tissues, it is norteworthy that cells can co-express two or more enzymes for nucleotide hydrolysis. As earlier described5, two enzymes are present on the ecto-surface of intact platelets: ecto-nucleoside triphosphate diphosphohydrolase (E-NTPDase) and ecto-5′-nucleotidase. In this study, we describe an ecto-nucleotide pyrophosphatase/phosphodiesterase (E-NPP) activity in rat platelets. Using p-nitrophenyl 5′-thymidine monophosphate (p-Nph-5′-TMP) as a substrate, we demonstrate an enzyme activity that shares the major biochemical properties described for E-NPPs: alkaline pH and divalent cation dependence, and blockade of activity by metal ion chelator. KM and Vmax values for p-Nph-5′-TMP hydrolysis were found to be 106 ± 18 mM and 3.44 ± 0.18 nmol p-nitrophenol/min/mg (mean±SD, n=5). We hypothesize that an E-NPP is co-localized with an ENTPDase and an ecto-5′-nucleotidase on the platelet surface. Thus, 0.25mM suramin inhibited p-Nph-5′-TMP, ATP and ADP hydrolysis, while 0.5mM AMP decreased only p-Nph-5′-TMP hydrolysis. Besides, 5.0, 10 and 20mM sodium azide just inhibited ATP and ADP hydrolysis. Angiotensin II (5.0 and 10nM) affected only ADP hydrolysis. Gadolinium chloride (0.2 and 0.5mM) strongly inhibited the ATP and ADP hydrolysis. The set of results presented here reinforces our hypothesis of the possible co-localization of E-NTPDase and E-NPP activities that, together with and ecto-5′-nucleotidase could constitute a multienzymatic complex for the complete nucleotide hydrolysis in the ecto-surface of intact rat platelets. The enormous tissue distribution of the members of the E-NTPDase and E-NPP families make it difficult to assign specific functions to these enzymes in individual tissues4. However, since these enzymes present different kinetic properties, they can act under distinct physiological conditions and can be differently regulated. By converting ADP, released from aggregated platelets to AMP, E-NTPDase may play an important role in the prevention of microthrombus formation5. In addition, phosphodiesterases may act as “guard dogs” to prevent subversion of the cell by destroying incoming DNA or RNA6. The E-NPP described here represents a novel insight into the control of platelet purinergic signaling.

Effect of Diadenosine Polyphosphates in Achondroplasic Chondrocytes: Inhibitory Effect of Ap4A on FGF9 induced MAPK cascade

Ana Isabel Guzmán, Marta Irazu and Jesús Pintor

Departamento de Bioquímica y Biología Molecular IV, E.U. Óptica, Universidad Complutense de Madrid, c/Arcos de Jalón s/n 28037 Madrid, Spain. aguzman@opt.ucm.es

Achondroplasia (dwarfism) is a genetic pathology due to a mutation in the gene that encodes for the fibroblast growth factor type 3 receptor in cartilage cells, the chondrocytes. Studies carried out in different populations have demonstrated that the most common mutation in achondroplasia is Gly380Arg position in the receptor transmembrane domain. This mutation produces a sustained activation of the receptor altering the normal equilibrium between proliferation and maturation therefore inhibiting normal growth of the bone (L’Hote et Knowles., 2005). Rat chondrocytes which have been transfected with the wild type form of human FGFR3 receptor and with the mutated type form, were used to analyse in depth the intracellular signal transduction pathways triggered by over-activated FGFR3. In this sense we have confirmed the induction of mitogen-activated protein kinases (MAP kinases) cascade. This pathway seems to be involved in alteration of chondrocyte differentiation process. Additionally this pathway elicits a lower synthesis of the extracellular matrix inhibiting bone growth (Yasoda et al., 2004).

Therefore a good aproache for the treatment of achondroplasia could be focused to limit the activity of the MAP kinase cascade, which is overstimulated by mutated FGFR3.

In order to modify the MAP kinase activity and taking into account that previously we have observed the presence of P2 receptors in achondroplasic chondrocytes, we have examined the effect of different dinucleotides and nucleotides on MAP kinase activation.

Diadenosine triphosphate, Ap3A and diadenosine pentaphosphate Ap5A did not modify phosphorylation of MAP kinase in achondroplasic chondrocytes. In contrast, diadenosine tetraphosphate, Ap4A reduced phosphorylation of MAP kinase. Other nucleotides assayed, both naturally occurring such as ATP, ADP, UTP and UDP or synthetic, increased or did not modify the levels of phosphorylated MAP kinase. This fact opens a new perspective inviting to consider Ap4A as a new therapeutic target for the treatment of achondroplasia.

This work has been supported by a research grant from la Comunidad de Madrid CAM GR/SAL/057372004, Fundación López Hidalgo and Fundación ICO.

Effect of The Neuroleptic Chlorpromazine on the Nucleotide and Phospholipid Metabolism in Human Lymphocytes.

Gergely Keszler, Rita Sümeg, Tanjana Spasokoukotskaja, Lynette D. Fairbanks,+ Annamaria Stenger*, H. Anne Simmonds+ and Maria Staub**

Semmelweis University, Medical Faculty, Department of Medical Chemistry, Molecular Biology and Pathobiochemistry 1444-Budapest, POB 260, Hungary * Health Service, Pediatric Department, District II, Budapest, Hungary +Purine Research Laboratory, United Medical and Dental Schools, Guy's Hospital, London, SE1 9RT, United Kingdom.

Chlorpromazine (CPZ) the amphiphilic amine, widely used as neuroleptic drug, decreased the ATP concentration by 30% in human lymphocytes, at relatively low concentration (30 µM) and short incubation time (60 min) without influencing the GTP, UTP and CTP concentrations. CPZ had the most pronounced effect on the nucleotide containing phospholipid precursors of the cells. At the same conditions, CDP-ethanolamine (CDP-EA) decreased from 87 to 75 pmol/million cells, while CDP-choline (CDP-Ch) from 12 to 0 pmol/million cells during incubation with 30 µM CPZ for 60 minutes. The effect of the anti-leukemic drug, arabinosyl-cytosine (araC) was also measured on the phospholipid precursor concentration, as it has been shown, deoxycytidine (dCyt) and its analogue araC can be also metabolized into the the phospholipids. AraC (50 nM) only slightly decreased the CDP-EA (from 87 to 85 pmol) concentration, while it doubled the CDP-Ch concentration (12 to 25 pmol) of the cells during 60 min incubation. The incorporation of labeled dCyt into dCDP-EA and dCDP-Ch were both inhibited by CPZ, while araC increased the labeling of dCDP-EA and slightly enhanced it in dCDP-Ch.

The inositol-phospholipid pathway can only be labelled from extracellular dCyt, if its turnover was inhibited by CPZ. Labelled dCyt appeared in dCDP-DAG only in the presence of CPZ, while its incorporation into DNA decreased.

Similar effect of CPZ on the phospholipid biosynthesis were found in Y79 neuroblastoma cell lines to that, measured in lymphocytes. Thus low concentrations of CPZ inhibited at many places the membrane phospholipid synthesis, which might be its primary effect on neural functions.

Effect of adenosine receptor activation on lipopolysaccharide (LPS)-evoked cytokine production in BV-2 microglia cells

Balázs Koscsó, Zsolt Selmeczy, E. Sylvester Vizi, György Haskó

Department of Pharmacology, Institute of Experimental Medicine, Hungarian Academy of Sciences, Budapest, Hungary koscso@koki.hu

Microglia, the intrinsic macrophages of the central nervous system, are normally quiescent, but rapidly respond to various kinds of noxious stimuli and these responses are relatively uniform irrespective of the noxius agent. One of these noxious agents is bacterial LPS, which induces elaboration of a wide array of proinflammatory mediators including cytokines and free radicals, as well as modulating the expression of cell surface markers. Although the sequence of intracellular events during microglial activation in response to LPS has been well defined, we have relatively little knowledge about the extracellular signals which affect the inflammatory processes of microglia. Adenosine is an endogenous purine nucleoside interacting with specific G protein-coupled receptors (A1, A2A, A2B, and A3), and is well recognized as an effective modulator of the immune system. In this study, we investigated whether adenosine receptors regulate production of the proinflammatory cytokine TNF-α in the BV-2 murine microglial cell line. BV-2 cells were treated with increasing concentrations of different adenosine receptor agonists (N6-benzyl 5′-N-ethylcarboxamidoadenosine (NECA), a non-selective adenosine receptor agonist; 2-chloro-N6-cyclopentyl-adenosine (CCPA), an A1 receptor agonist; CGS-21680, an A2A receptor agonist; and N6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (IB-MECA), an A3 receptor agonist) 30 minutes before addition of 10 µg/ml LPS. 24 h later, the supernatants were collected and analyzed for TNF-α content using ELISA. Our results showed that pretreatment with adenosine receptor agonists decreased LPS-induced TNF-α production in BV-2 cells in the following order of potency: IB-MECA≥CCPA>NECA>CGS-21680, indicating a primary role for A3 receptors. Next, the cells were pretreated with various selective adenosine receptor antagonists (8-cyclopentyl-1,3-dipropylxanthine (DPCPX), A1 antagonist; 4-(2-[7-amino-2-(2-furyl)[1,2,4] triazolo [2,3-a] [1,3,5] triazin-5-ylamino]ethyl)phenol (ZM241385), an A2A antagonist; benzo[g]pteridine-2,4(1H,3H)-dione isoalloxazine (alloxazine), an A2B receptor antagonist; and 5-propyl-2-ethyl-4-propyl-3-(ethylsulfanylcarbonyl)-6-phenylpyridine-5-carboxylate (MRS-1523), an A3 receptor antagonist) before treatment with CCPA or IB-MECA. MRS-1523, but not other antagonists, reversed the suppressive effects of both IB-MECA and CCPA on LPS-induced TNF-α production. These results demonstrate that A3 receptor stimulation suppresses TNF-α production by BV-2 microglia. Thus, adenosine may be an important endogenous anti-inflammatory mediator in the brain.

Effect of low frequency electromagnetic fields on A2A and A3 adenosine receptors in human neutrophils

Katia Varani1, Stefania Gessi1, Stefania Merighi1, Fabrizio Vincenzi1, Elena Cattabriga1, Annalisa Benini1, Ruggero Cadossi2, Pier Andrea Borea1

1Department of Clinical and Experimental Medicine, Pharmacology Unit, University of Ferrara, Ferrara, Italy, 2 Laboratory of Biophysics, IGEA, Carpi, Italy

The present study was designed to evaluate the binding and functional characterization of A2A and A3 adenosine receptors in human neutrophils exposed to low frequency, low energy, pulsing electromagnetic fields (PEMFs). Saturation binding experiments on A2A and A3 adenosine receptors revealed a single class of binding sites with similar affinity in control and in PEMF-treated human neutrophils (KD=1.05±0.10 and 1.08±0.12 nM for A2A receptors and KD=2.36±0.16 and 2.45±0.15 nM for A3 receptors, respectively). PEMFs treatment revealed that the A2A and A3 receptor density was statistically increased (Bmax=126±10 and 215±15* fmol/mg protein and Bmax= 451±18 and 736±25* fmol/mg protein, respectively,*, P<0.01). In the adenylyl cyclase assays the A2A agonists, HE-NECA and NECA increased cAMP accumulation in untreated human neutrophils with an EC50 value of 43 (40–47) and 255 (228–284) nM, respectively. The capability of the same compounds to stimulate cAMP levels in PEMF treated human neutrophils was increased with an EC50 value of 10 (8–13)* and 61 (52–71)* nM, respectively, *, P<0.01. In the superoxide anion production assays examined agonists inhibited the generation of O2 in untreated human neutrophils with an EC50 value of 3.6 (3.1–4.2) and of 23 (20–27) nM, respectively. In PEMF treated human neutrophils, the same compounds show an EC50 value of 1.6 (1.2–2.1)* and of 6.0 (4.7–7.5)* nM, respectively, *, P<0.01. Similarly, typical A3 agonists such as Cl-IB-MECA and IB-MECA were able to inhibit cyclic AMP accumulation and superoxide anion production. The potencies of Cl-IB-MECA and IB-MECA were statistically increased after exposure to PEMFs. These data indicate in human neutrophils treated with PEMFs the presence of significant alterations in the A2A and A3 adenosine receptor density and functionality. In summary, these results should serve as an impetus for further investigation of the pharmacological changes in the adenosine receptors and PEMF treatment. From a clinical point of view a clarification of the potential effects of PEMFs facilitate the development of alternative treatments or the elaboration of novel promising therapeutic tools.

Effect of P2 receptor antagonists on human ecto-nucleotidases.

M.N. Munkonda, S.A. Lévesque, F. Kukulski, C. Legendre, J. Lecka and J. Sévigny.

Centre de recherche en Rhumatologie-Immunologie du CHUL, Département de médecine, Université Laval, Québec, Canada. sebastien.levesque@crchul.ulaval.ca


Through the activation of P2-receptors, extracellular nucleotides regulate a variety of biological functions. Concentrations of extracellular nucleotides are modulated by ectonucleotidases such as by members of the ecto-nucleoside triphosphate diphosphohydrolase (E-NTPDase) and ecto-nucleotide pyrophosphatases/phosphodiesterases (E-NPPs) families. These enzymes hydrolyze the terminal phosphate residue of tri- and/or diphosphonucleosides, and, in concert with ecto-5′-nucleotidase, facilitates the formation of adenosine1. Four different NTPDases, bound to plasma membrane, appears important in the control of nucleotide concentration, namely NTPDase1 (CD39), NTPDase2, 3 and 8.2 Also, NPP1 and NPP3 could be involved in the control of extracellular nucleotides concentrations.1 The lack of specific inhibitors for these enzymes impedes the research on their functions. It has already been reported that several P2 receptor antagonists such as Reactive Blue-2 (RB-2) and suramin also inhibit ecto-nucleotidases.3 In this study, we investigated the effect of several P2-receptor antagonists (suramin, NF279, NF449, RB-2 and MRS2179) on the activity of human NTPDases 1, 2, 3 and 8 and human NPP1 and NPP3.


Recombinant human NTPDases and NPPs were produced by transfecting COS-7 or HEK 293T cells with an expression vector encoding each of these enzymes. The inhibition of NTPDase activity (ATP and ADP hydrolysis) was measured by the quantification of released phosphate (Pi). The inhibition NPPs activity (pnp-TMP hydrolysis) was measured by the quantification of para-nitrophenol produce using a colorimetric assay.


Most NTPDases were inhibited by the P2-receptor antagonists tested. Broad range antagonists RB-2 (100 µM) blocked all NTPDases activity by over 87% while suramin (100 µM) inhibited mainly NTPDases 2 and 3 to 51±2% and 66±5% respectively. RB-2 completely inhibited human NPP1 and decreased activity of human NPP3 by about 40% while suramin inhibited these enzymes by 70±2% and 8±3%, respectively. More specific antagonists like suramin derivative NF279 (P2X1, P2X7), NF449 (P2X1) blocked NTPDases 1 to 3 by over 70% at concentration of 100 µM while MRS2179 (P2Y1) only affected human NTPDase3 by about 50%. Interestingly, NTPDase3 could be blocked by NF279 and NF449 at lower concentrations with EC50 of 0,1 and 5 2M respectively. All these specific antagonists blocked human NPP1 activity by over 75% while they had minimal effect on human NPP3 (<15%). These data on P2-receptor antagonists are relevant for the interpretation of pharmacological experiments with these molecules. Some of these P2 receptor antagonists may also be used as inhibitors of ecto-nucleotidases and even discriminate between NTPDase3 activity and the other ecto-nucleotidases.

Effect of Sulphate-Reducing Bacteria Infection on ATP-induced Apoptosis of Human Intestinal Epithelial Cells.

Carolina de Oliveira Souza1,2, Robson Coutinho-Silva1, Maurício Magalhães de Paiva2, and Cláudia Mara Lara Melo Coutinho2,3

1Instituto de Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Brasil. 2Instituto Oswaldo Cruz, FIOCRUZ, Rio de Janeiro, RJ, Brasil. 3Instituto de Biologia, Universidade Federal Fluminense, Niterói, RJ, Brasil. ccoutinho@ioc.fiocruz.br

Sulphate-reducing bacteria (SRB) are a group of anaerobic organisms involved in the reduction of sulphate to sulphide and are present in the majority of natural environments, plants, animals and humans. SRB were found colonizing the human gut and have been associated with the development of ulcerative colitis, a kind of chronic inflammatory bowel disease (IBD). Recently, it was shown that nucleoside-nucleotide free diet protects rat colonic mucosa from damage in a model of IBD. It was also shown that P2X and P2Y subtype receptors can be positively modulated in smooth muscle and immune cells, respectively, on human IBD. Here, we evaluated if the interaction of SRB with human intestinal epithelial cells can interfere with ATP-induced apoptosis on these cells.

Material and Methods

a subconfluent culture of HCT8 human intestinal epithelial cells (106 total cells) was infected or not with pure SRB strain Desulfovibrio indonensiensis (final concentration of 106 cells.ml−1) and, after 24 h of infection, the cells were washed and treated with 2mM of ATP for 12 to 48 h. The attached cells were then released from the wells with PBS/EDTA and recovered in Falcon tubes, washed and centrifuged for 7 min at 250g. The cells were finally resuspended in apoptosis buffer containing 0.1% sodium citrate, 0.1% Triton X-100 and 5 g/ml of propidium iodate. The DNA content of the acquired cells (10.000 cells per tube) was analyzed using a FACScan Becton Dickinson flow cytometry. SRB-infected and control epithelial cells were processed for scanning electron microscopy analysis.


we observed that both ATP and SRB induced specific apoptosis on 11 ± 2 % and 6 ± 1 % of HCT8 cells, respectively. When SRB were led to infect the cells followed by ATP treatment, we observed stimulatory effect on ATP-induced apoptosis growing up to 19 ±2 % of cells. In addition, scanning electron microscopy analyses showed that SBR were found attached to the surface of the epithelial cells. It was also observed significant differences on the morphological pattern of the infected epithelial cells, especially increased number of microvilla projecting from their plasma membranes.


Our data suggest that SRB attach and cause apoptosis to intestinal epithelial cells and such bacterial infection can interfere with P2 receptor signaling of these epithelial cells. We hypothesize that SRB together with P2 receptor can have a role on the establishment of IBD.



Effector-coupling of the A2A adenosine receptor is contingent on the action of retinoic acid in neuroblastoma cells

Edin Ibrisimovic, Helmut Drobny, Stefan Boehm, Eduard Stefan, Christian Nanoff and Oliver Kudlacek

Institute of Pharmacology, Center for Biomolecular Medicine and Pharmacology, Medical University of Vienna, Austria edin.ibrisimovic@meduniwien.ac.at

The A2A-adenosine receptor has been implicated in the pathogenesis of movement disorders, as a drug target it holds therapeutic potential. We have investigated if SH-SY5Y neuroblastoma cells (SH-cells) which endogenously express an A2A-receptor and can be differentiated into a neuron-like phenotype may serve as a model for the regulation of A2A-receptor signalling in nerve cells. We find that receptor activation facilitated neurotransmitter release and that the receptor molecule was targeted to cell extensions reflecting previous evidence obtained on brain slices. An additional finding in SH-cells was that coupling of the receptor to its prototypical effector adenylyl cyclase is contingent on pre-treatment of cells with all-trans retinoic acid (RA). We have investigated if changes in the signal transduction complement accounted for the enhanced coupling. Productive receptor-effector coupling correlated with an increase in catalyst activity but the treatment with RA had a stronger effect on receptor- (∼20-fold) than on forskolin-mediated cAMP formation (∼4-fold). The maximum effect occurred one day after pre-treatment of cells with RA. Both the time course and the regulation pattern consistent with the presence of type I and V isoforms were suggestive of transcriptional regulation via the RA-receptor. However, we did not detect alterations in the mRNA level specific for these and other adenylyl cyclase isoform transcripts in SH-cells (type I, III, V, VII, IX) nor in the amount of membrane bound adenylyl cyclase type I and type V protein. In G-protein levels we found a ∼20% increase in Gαs with no changes in Gαi/o or Gβγ. Nevertheless, the increase could hardly explain the increment in activity nor the altered regulatory pattern. The A2A-receptor level was constantly low (∼100 fmol/mg) even upon the phenotypic changes induced by treatment with RA. Our current hypothesis is that RA fortifies coupling by an indirect effect via an autocrine regulatory mechanism. This is based on the following evidence. RA is known to induce the BDNF receptor trkB in SH-cells and we detected BDNF transcript in SH-cells treated with RA. Conversely, addition of a trkB inhibitor (K252a) during the RAdependent maturation interval impaired the effect, i.e. coupling of adenylyl cyclase to the A2A-receptor. This raises the issue if RA might affect A2A-signalling in the adult brain.

Effects of adenosine receptor agonists and antagonists in WAG/Rij rats, a genetic model of generalized absence epilepsy.

1Citraro R., 1Ferreri G., 1Russo E., 2D'Auro M., 2Ciccarelli R., 2De Sarro G.

1Chair of Pharmacology, 2Department of Experimental and Clinical Medicine, “Magna Graecia” University. Catanzaro. Italy. 2Department of Biomedical Sciences, University of Chieti-Pescara. Chieti. Italy. r.ciccarelli@dsb.unich.it

Adenosine exerts anticonvulsant effects in convulsant epilepsies, mainly via inhibitory A1 receptors, whereas the pharmacological blockade or genetic inactivation of the facilitatory A2A subtypes seems to attenuate convulsions1. However, the effects of A1 and A2A receptor activation were not yet evaluated in a particular type of epilepsy such as absence seizures, characterised by brief interruption of consciousness associated with generalised, synchronous spike-wave discharges (SWDs) in the EEG. Thus, we wanted to test whether the focal bilateral microinjection of A1 and A2A selective adenosine receptor agonists and antagonists into thalamic nuclei (nucleus reticularis thalami, NRT; ventroposterolateral, VPL; ventroposteromedial, VPM) and the peri-oral region of the somatosensory cortex (S1po), brain areas mainly involved in absence seizure triggering, modified the number and/or duration of SWDs in a genetic animal model of absence epilepsy, the WAG/Rij rat. In animals equipped with fronto-parietal cortical electrodes for EEGraphic recordings and two additional guide cannulae for bilateral focal microinjection, drugs were focally administered in a volume of 0.5 µl/side2. We evaluated the effects of drugs selectively active on A1 (agonist= CCPA; antagonist= DPCPX) and A2A receptors (agonists= CGS21680 and 2HE-NECA; antagonist= SCH58261). Independently of the site of administration, CCPA dose-dependently reduced the number of SWDs, while affected their duration differently, i.e. CCPA increased or decreased SWD duration when microinjected in VPM or VPL, respectively, whereas duration was not significantly modified in the other areas examined. The A2A agonists had no effects in the VPL, reduced the number and increased SWD duration in the VPM and S1po, while, in the NRT, they increased SWD number without affecting the duration. The A1 antagonist dose-dependently reduced the SWD number in all the examined areas, without affecting the duration, except in NRT, where the duration was significantly increased. Similarly, the A2A antagonist significantly reduced SWD number in all injected areas, modifying their duration only in NRT (increase) and VPL (reduction). In conclusion, these results are mostly in agreement with a few previous findings indicating either a protective activity by non selective adenosine receptor blockade with theophylline3 or a pro-convulsive adenosine4 effect in the same animal model of absence epilepsy. Further studies need to better evaluate whether, in this kind of epilepsy, is prevailing the pro-convulsant A1-mediated effect, probably related to an increased inhibitory function in the brain, which plays a significant role in the absence seizure pathogenesis, or the pro-convulsant activity of the excitatory A2A receptors.

Effects Of Adenosine Receptor Antagonists On Amitriptyline-Induced QRS Prolongation In Isolated Rat Hearts

Akgun A, Kalkan S, Hocaoglu N, Gidener S, Tuncok Y.

Dokuz Eylul University, School of Medicine, Department of Pharmacology, Izmir, Turkey. aylin.akgun@deu.edu.tr


In our previous study, we had demonstrated that adenosine A1 and A2a antagonists prevented hypotension and QRS prolongation in an in-vivo model of rat amitriptyline toxicity1. The aim of the study was to investigate the effects of adenosine receptor antagonists on amitriptyline-induced cardiotoxicity in isolated rat hearts.


A randomized controlled experimental study was performed with hearts of adult male Wistar rats mounted on a cannula of the Langendorff isolated perfused heart apparatus with constant flow. Left Ventricular Developed Pressure (LVDP), dp/dtmax, QRS duration and Heart Rate (HR) were measured. Two experimental protocols were performed. The amitriptyline concentration that prolongated the QRS duration more than 150 % (10−4 M) was accepted as the control group for the first protocol and the concentration that prolongated the QRS duration 50–75 % (10−4.25M) was accepted as the control group for the second protocol. In the first protocol, 10−4 M amitriptyline was infused for 60 minutes following pretreatment with a selective adenosine A1 receptor antagonist, DPCPX (8-cyclopentyl-1,3-Dipropylxanthine, 10−4 to 10−6 M) or a selective adenosine A2a receptor antagonist, CSC (8-3-chlorostyryl -caffeine,10−4 to 10−6 M). At the second protocol, 10−4.25 M amitriptyline was infused for 60 minutes following pretreatment with DPCPX (10−4 M) or CSC (10−5 M). Statistical analyses were performed by Student's t test for paired data and ANOVA followed by Tukey-Kramer for multiple comparison tests.


At the first protocol, 10−4 M amitriptyline infusion following pretreatment with 10−4 M DPCPX shortened QRS duration at 50 min. significantly when compared to control group (146.7±13.3 % for DPCPX pretreatment, 275.0±40.3 % for control, p<0.05). Amitriptyline infusion following pretreatment with 10−4 to 10−6 M DPCPX or 10−5 or 10−6 M CSC did not change LVDP, dp/dtmax and HR when compared to control (p>0.05). At the second protocol, pretreatment with 10−4 M DPCPX shortened the QRS duration (122.2±7.0 % at 40 min., p<0.05; 120.8±6.7 % at 50 min., p<0.01;120.8±6.7 % at 60 min., p<0.05). Pretreatment with 10−5 M CSC prolongated QRS duration (185.4±20.8 % at 20 min., p<0.05; 187.5±12.5 % at 30 min., p<0.05; 218.8±14.6 % at 60 min., p<0.05). Amitriptyline infusion (10−4.25 M) following pretreatment with 10−4 M DPCPX or 10−5 M CSC did not change LVDP, dp/dtmax and HR when compared to control (p>0.05).


While 10−4 M DPCPX shortened QRS prolongation, 10−5 M CSC prolongated QRS duration in the isolated rat hearts with prolonged QRS by 50–75 % induced by 10−4.25 M amitriptyline. However, 10−4 M DPCPX was not sufficient to shorten QRS prolongation induced by the higher concentration of amitriptyline(10−4 M). According to our results, adenosine A1 receptors might have a role in amitriptyline-induced QRS prolongation. Further studies with electrophysiological effects of adenosine antagonists on action potential for amitriptyline cardiotoxicity is needed to explain the exact mechanism.

Effects of adenosine-angiotensin II interactions on renal function

M. Morato1,2, J. Marques-Lopes1, T. Sousa1, D. Pinho1, M. Couto1, S. Fontes1, A. Albino-Teixeira1

1Institute of Pharmacology and Therapeutics, Faculty of Medicine of Porto and IBMC; 2Department of Pharmacology, Faculty of Pharmacy of Porto; University of Porto, Portugal; mmorato@ff.up.pt

Selective antagonists of adenosine A1 but not A2 receptors produce diuresis and natriuresis (1,2). Non-selective antagonism of adenosine receptors with 1,3-dipropyl-8-sulfophenylxanthine (DPSPX) induces hypertension (3) and activation of the renin-angiotensin system (4,5). Angiotensin II induces tubular sodium reabsorption (6). This study aimed at assessing renal function in DPSPX-hypertensive rats.

DPSPX (90 µg/kg/h; i.p.) was continuously infused from day 0 to day 7 to male Wistar rats (240–270g). Shamoperated (S; n=6) and DPSPX (PX; n=5)-treated rats were housed in metabolic cages from day -2 to day 14 with free access to water and rat chow. Intakes were monitored every day. Twenty-four hours urine was daily collected. Plasma was collected on day 14. Sodium and potassium levels were quantified with ion-sensitive electrodes. Urea was measured by an enzymatic test, creatinine by the Jaffé method and total proteins by the biuret reaction. Systolic blood pressure (SBP) was determined by tail-cuff. Statistical analysis was performed using Student's t test. DPSPX increased SBP (S=123.20±1.13 mmHg; PX=147.57±5.83 mmHg; p<0.05). Infusion of DPSPX decreased food and water intake until day 7 and urinary excretion rate throughout the study (Table). Body weight gain was similar between S and PX groups. DPSPX-induced hypertension was associated with reduced urinary sodium, potassium and urea excretion rates (Table). No difference was found in urinary creatinine or protein excretion between S and PX groups. On day 14, there was no difference in plasma sodium, potassium, urea, creatinine or protein levels between S and PX groups. Creatinine clearance was similar between S and PX groups. Fraccional excretion of potassium was lower in DPSPX-hypertensive rats (S=0.6±0.06 %; PX=0.4±0.02 %; p<0.05) while fraccional excretion of sodium was similar to that of sham-operated rats.

Day 7


Day 13






Food intake (g)





Water intake (mL)





Uexcretion rate (mL/min)

10.8 ±0.77




UNa+ (mg/d/'kg)





UK+ (mg/d/kg)





Uurea (g/d/kg)





We concluded that DPSPX-induced hypertension is associated with decreased excretion of fluid, electrolyte and urea. Future studies will clarify the net result of the concomitant antagonism of adenosine receptors with increased plasma angiotensin II levels on renal function.

Supported by POCTI/NSE/45409/2002 and POCTI/NSE/38952/2001 of FCT and FEDER

Effects of exogenous cAMP on rat intestinal motility

Maria Cecilia Giron, Anna Bin, Elisabetta Ciervo, Chiara Zoppellaro, Federica Bianchi, Rosa Maria Gaion

Department of Pharmacology and Anesthesiology, University of Padova cecilia.giron@unipd.it


In these last few decades it has been proposed that cyclic AMP (cAMP) could be an extracellular mediator, supported by the following facts: in several cell types cAMP was shown to be extruded into interstitial fluids and to bind to specific plasma membrane sites. Moreover extracellular cAMP can affect specific functions in a variety of cells and tissues1,2. AIMS: The purpose of the present investigation was to evaluate the action of exogenous cAMP on the isolated rat ileum and to determine whether any such action is a) affected by inhibition of tissue phosphodiesterase or ecto-phosphodiesterase or ecto-5′nucleotidase activity, b) mediated by P1 receptors and/or release of endogenous acetylcholine, c) possibly governed by the adrenergic activity.


To measure the muscle tension, ileum segments (approximately 2 cm in length, 0.2±0.03 g wet tissue weight) from Wistar rats (250±50 g b. w.) were mounted vertically in double-jacketed organ baths containing 10 mL of Tyrode solution (NaCl 136 mM, KCl 2.7 mM, CaCl2 1.4 mM, MgCl2 0.46 mM, NaH2PO4 0.32 mM, NaHCO3 12 mM, glucose 5 mM), maintained at 37°C and continuously aerated with a mixture of 95% O2 and 5% CO2. Changes in tension were recorded by isometric transducers coupled to a pen recorder, and they were expressed as a percentage of the maximal response to carbachol (1 µM) measured at the beginning of each experiment.


Addition of cAMP (0.1 µM to 1 mM) evoked a slight, transient relaxation which, in most cases (90%), was followed by a marked, concentration-dependent contraction. Contractions began within 30 s after application to the bath and the peak response was reached within 60 s. cAMP-induced contraction was sustained (over 15 min) and reversible with washing. Repeated administration of 0.5 mM cAMP, followed by washing at 15 min intervals, did not induce tachyphylaxis. Pretreatments with 3-isobutyl-1-methylxanthine (IBMX, 10 µM), a PDE-inhibitor, 1,3-dipropyl-8-p-sulfophenylxanthine (DPSPX, 0.1 mM), an ecto-PDE-inhibitor, α,β methylene-ADP (AOPCP, 0.1 mM), an ecto-5′NT-inhibitor, partially attenuated cAMP-induced contractions by 29%, 37% and 50%, respectively. Pretreatment of the preparations with 8-phenyltheophylline (8-PT, 10 µM), a P1 antagonist, or with atropine (1 2M), a muscarinic antagonist, did not significantly influence cAMP effect on ileum tension. Concentration-dependent contractions induced by cAMP were significantly reduced by in vivo treatment of rats with reserpine, to deplete norepinephrine (NE) stores in nerve terminals.


These experimental observations suggest that extracellular cAMP may act directly on intestinal smooth muscle by binding to specific cell membrane sites1,3, but degradation products of the cyclic nucleotide are also involved in its effects on rat ileum, that are controlled by the adrenergic system.

Effects of genetic deletion of adenosine deaminase and A1 receptors in normoxic and ischemic hearts

Melissa Reichelt*1, Laura Willems*1, Jose G. Molina2, Chun-Xiao Sun2, Janci C. Noble2, Kevin J. Ashton1, Jurgen Schnermann3, Michael R. Blackburn2, and John P. Headrick1†

*Equal first authorship

1Heart Foundation Research Center, Griffith University Southport, QLD 4217, Australia, 2Department of Biochemistry and Molecular Biology, University of Texas Health Science Center at Houston, Medical School, Houston, TX 77030, USA, and 3National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA m.reichelt@griffith.edu.au

Adenosine deaminase (ADA) may be multifunctional,regulating adenosine levels and receptor (AR) agonism and functionality. We assessed effects of ADA, A1AR, and dualADA/A1AR knockout (KO) on AR-mediated responses and ischemic tolerance (20-25 min ischemia 45 min reperfusion) in murine hearts. Neither ADA or A1AR KO modified basal contractility, though ADA KO reduced resting heart rate (an effect abrogated by A1AR KO). AR mediated bradycardia and dilation with 2-chloroadenosine was unaltered by ADA KO, and A1AR KO eliminated bradycardia. Adenosine efflux was increased 10- to 20-fold by ADA KO (at the expense of inosine). Deletion of ADA improved outcome from 25 min ischemia, reducing diastolic pressure (21±4 vs. 38±3 mmHg) and LDH efflux (0.12T0.01 vs. 0.21T0.02 U/g/min ischemia), and enhancing pressure development (89±6 vs. 66±5 mmHg). Protection was also evident after 20 min ischemia, and mimicked by the ADA inhibitor EHNA (5 µM). ADA KO enhanced tolerance in A1AR KO hearts, though effects on diastolic function were eliminated. Absence of ADA does not alter functional sensitivities of cardiovascular A1 or A2ARs, despite enhanced adenosine levels, but enhances ischemic tolerance. Conversely, A1AR KO impairs ischemic tolerance. Effects of ADA KO on diastolic dysfunction are A1AR-specific while other ARs contribute to changes in contractile recovery and cell death.

Effects of the Adenosine Receptor Antagonists on Amitriptyline-Induced Vasodilation in Rat Isolated Aorta

Kalkan S, Hocaoglu N, Akgun A, Gidener S, Tuncok Y.

Dokuz Eylul University, School of Medicine, Department of Pharmacology, Izmir, Turkey. sule.kalkan@deu.edu.tr


Although, we had demonstrated that adenosine receptor antagonists prevented hypotension in an invivo rat model of amitriptyline toxicity1, it was not clear that whether adenosine receptors in heart or in vasculature were dominant. The aim of the study was to investigate the role of adenosine A2a receptors on amitriptyline-induced vasodilation in rat isolated aorta.


After determining EC80 of noradrenalin (the concentration of noradrenalin that produces 80% of maximal contractile response) as 10−5M, IC50 value of amitriptyline was calculated as the drug concentration causing a half-maximal inhibition of contractile responses to noradrenalin (NA) in rat isolated aorta. In the experimental groups, tissues were first contracted with EC80 of NA, then, DPCPX (8-cyclopentyl-1,3-Dipropylxantine, a selective adenosine A1 antagonist, 10−9–10−5 M), CSC (8-(3-chlorostyryl), a selective A2a antagonist, 10−9-10−5 M) or DMSO (dimethyl sulfoxide, a solvent for adenosine antagonists) were incubated. Following the amitriptyline incubation, NA was administered again. IC50 of amitriptyline was compared in the presence of the DPCPX, CSC or DMSO. Statistical analysis was done by Student's t test.


Amitriptyline inhibited 49.9 ± 3.7 % contractile response to NA on aorta segments at 1.8 ×10−5M (IC50). DPCPX increased amitriptyline-induced inhibition on contractile response to NA dose dependently. CSC decreased amitriptyline- induced inhibition on contractile response to NA at 10−5M. DMSO did not change amitriptyline- induced inhibition on contractile response to NA, significantly.


Adenosine A2a receptor stimulation seems to be responsible for amitriptyline-induced vasodilation and hypotension since adenosine A1 antagonist, DPCPX, increased amitriptyline-induced.

Effects of the membrane lipid environment on adenosine A1 and A2A receptors in rat brain striatum

A. Themann, A. C. Schiedel and C. E. Müller

Pharmaceutical Sciences Bonn (PSB), Pharmaceutical Chemistry, Institute of Pharmacy, University of Bonn, Kreuzbergweg 26, 53115 Bonn, Germany andrea.themann@uni-bonn.de

Lipid rafts are plasma membrane microdomains rich in cholesterol and sphingolipids, which provide a particularly ordered lipid environment. They play a central role in many cellular processes, including membrane sorting and trafficking, cell polarization, and signal transduction processes. Certain G protein-coupled receptors (GPCRs) and associated molecules were shown to be enriched in cholesterol-rich microdomains which may act as signalling platforms [1]. The family of adenosine receptors, which are GPCRs, consists of four subtypes designated A1, A2A, A2B and A3. Adenosine A1 and A2A receptors are expressed in high density in the striatum (caudate-putamen). Striatal A2A receptors have been identified as novel targets for the treatment of Parkinson's disease and related motoric disorders (A2A antagonists) and for schizophrenia (A2A agonists) [2,3]. Lipid rafts were described as detergent-insoluble membranes (DRMs) that can be isolated by sucrose-gradient centrifugation after treatment of membranes with the detergent Triton X-100 at 4°C [4]. When striatal membranes were subjected to 0.4 % Triton X-100 followed by sucrose-gradient centrifugation, we found that A1 and A2A receptors were highly enriched in DRMs as shown by radioligand binding studies using the A1 agonist radioligand [3H]CCPA, and the A2A agonist radioligand [3H]CGS21680, respectively. However, we could also demonstrate that DRMs are detergent-induced artefacts, which are not representative of the structures in an intact cell membrane.

To examine the modulatory role of cholesterol on ligand binding and G-protein coupling of adenosine A1 and A2A receptors, cholesterol was depleted from rat brain striatal membranes using methyl-β-cyclodextrin (MβCD) [5]. Saturation binding experiments with the A1 agonist [3H]CCPA, the A1 antagonist [3H]DPCPX, the A2A agonist [3H]CGS21680 and the A2A antagonist [3H]ZM241385 were performed at native and cholesterol-depleted membranes. Cholesterol depletion significantly altered the affinity of A1 (increase) and A2A agonists (decrease) without affecting antagonist affinities. In addition, it reduced the number of receptors (Bmax values) recognized by agonists and antagonists. Furthermore, [35S]GTPγS binding assays were performed to assess receptor function. Our data provide evidence that the cholesterol content influences the interaction of ligands with the highly expressed adenosine A1 and A2A receptors in rat brain striatum. These results may have relevance for statintreated patients, and in Alzheimer's disease where enhanced cholesterol levels have been observed.

Electrophysiological characterization of aminoglycoside antibiotics effects on the P2X2 receptor and a P2X2/P2X1 receptor chimera after heterologous expression in Xenopus oocytes

Eva-Verena Bongartz and Jürgen Rettinger

Max-Planck-Institute of Biophysics, Frankfurt am Main, Germany Eva.Bongartz@mpibp-frankfurt.mpg.de

P2X receptors are cation-selective ligand-gated ion channels that open upon extracellular binding of ATP. Aminoglycoside antibiotics work by binding to the 30S ribosomal subunit of Gram-negative bacteria, causing inhibition of protein synthesis. Their clinical use is limited by toxic side effects that include cochlear and vestibular toxicity as well as nephrotoxicity. Micromolar concentrations of the aminoglycosides streptomycin or gentamycin are commonly used in the culture medium of Xenopus laevis oocytes. It has previously been described that neomycin causes voltage dependent inhibition of ATP-induced whole cell current of Guinea-pig outer hair cells (OHC) (1) which have later been shown to express P2X2 receptors. To further characterize aminoglycoside inhibition at defined P2X receptors and determine their mode of action we investigated the inhibitory potency of five different aminoglycoside antibiotics at homomeric rat P2X2 receptors and at a non-desensitizing P2X2/P2X1 receptor chimera which exhibits the ligand-binding properties of the P2X1 receptor (2). The respective subunits were expressed in Xenopus laevis oocytes and the ATP induced membrane currents were analysed using the twoelectrode voltage-clamp technique. At −60 mV, all five aminoglycosides inhibited ATP induced inward currents with varying potencies. The inhibition was dose dependent and displayed fast on-set and off-set. P2X2 receptor currents were most potently blocked by streptomycin with an IC50 value of 35 µM, the weakest inhibition was induced by paromomycin with half-maximum inhibition at 1.5 mM. The rank order of potency at P2X2 receptors was streptomycin > neomycin > gentamycin > kanamycin > paromomycin. This rank order differs from the results obtained from native P2X receptors on OHCs where neomycin was significantly more effective than streptomycin and gentamycin. Similar to what has been found on OHCs, the inhibition was voltage-dependent and greatly reduced at more positive membrane potentials. Likewise, the ATP-current could not be reversed by increasing the ATP concentration, indicating that the effect was non-competitive. To further analyze the blocking mechanism we tested the aminoglycosides on a non-desensitizing P2X2/P2X1 receptor chimera in which the N-terminal and first transmembrane portion of the P2X1 were replaced by the corresponding regions of the P2X2 receptor subunit and compared it to the properties of known P2X receptor antagonists. The onset rates of inhibition of PPNDS, NF449 or TNP-ATP at this chimera were limited by the slow dissociation rate of ATP (τ = 63 s). In contrast, all tested aminoglycosides displayed a fast onset of inhibition at the chimera thus clearly indicating an inhibitory action independent of ATP dissociation. Taken together, we show that aminoglycosides block the rat P2X2 receptor with a rank order of potency different from that at native P2X receptors on guinea-pig outer hair cells, and that the P2X2/P2X1 chimera provides a fast screening model to distinguish between competitive and noncompetitive antagonism.

Endothelial Catabolism of Extracellular Adenosine during Hypoxia: Role of Surface Adenosine Deaminase and CD26

Holger K. Eltzschig,1 Marion Faigle,1 Simone Knapp,1 Jorn Karhausen,1 Juan Ibla,2, 4 Kirsten C. Odegard,2 Peter C. Laussen,2 Linda F. Thompson,3 and Sean P. Colgan4

1Department of Anesthesiology and Intensive Care Medicine, Tübingen University Hospital, D-72076, Tübingen, Germany and 2Department of Anesthesiology, Perioperative and Pain Medicine, Children's Hospital, Harvard Medical School, Boston, Massachusetts, USA, and 3Immunobiology and Cancer Program, Oklahoma Medical Research Foundation (OMRF), Oklahoma City, OK 73104 and 4Center for Experimental Therapeutics and Reperfusion Injury, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115, USA heltzschig@partners.org

Extracellular levels of adenosine significantly increase during conditions of limited oxygen availability (hypoxia) (1, 2, 3). Therefore, we reasoned that adenosine clearance mechanisms must exist to counter potentially deleterious influences of adenosine (4). As guided by microarray results revealing a nearly fifty-fold induction of endothelial adenosine deaminase (ADA) mRNA in hypoxia, we pursued the hypothesis that hypoxia coordinates adaptive catabolism of adenosine. Utilizing in vitro and in vivo models of adenosine signaling, we confirmed induction of ADA at the protein and functional levels. Further studies revealed that the ADA complexing protein CD26 is coordinately induced by hypoxia, effectively localizing ADA activity at the endothelial cell surface. Functional studies in murine hypoxia models revealed that inhibition of ADA with coformycin enhances protective physiologic responses mediated by adenosine (e.g. vascular leak, neutrophil accumulation, pulmonary edema). Analysis of plasma ADA activity in pediatric patients with chronic hypoxia (mean oxygen saturation of 81.9±6%) undergoing cardiac surgery (Bi-directional Glenn procedure) revealed a 4.1±0.6-fold increase in plasma ADA activity compared to controls. Taken together, these results demonstrate induction of ADA as an innate metabolic adaptation to elevated extracellular adenosine levels during hypoxia, and identify ADA inhibition as potential therapeutic strategy for vascular leakage and excessive inflammation associated with acute hypoxia.

Endothelium-dependent adenosine A2B-receptor-mediated relaxation of corpora cavernosa is impaired in men with vasculogenic impotence

M. Faria1, M.T. Magalhães-Cardoso1, J.-M. Lafuente-de-Carvalho2 & P. Correia-de-Sá1

1Laboratório de Farmacologia, Unidade Multidisciplinar de Investigação Biomédica (UMIB), and 2Serviço de Urologia, Hospital Geral de Santo António (HGSA), Instituto de Ciências Biomédicas de Abel Salazar (ICBAS) — Universidade do Porto, Portugal. (migfaria@icbas.up.pt)

Purinergic transmission is important for initiation and maintenance of penile erection. Smooth muscle relaxation to ATP may be mediated directly (via P2-purinoceptors) or indirectly, via adenosine generated by ectonucleotidases (Mantelli et al., 1995, J. Androl., 16, 312–317). Due to its short half-life, adenosine has been used as an agent for the diagnosis of vasculogenic impotence (Kilic et al., 1994, Int. J. Impot. Res., 6, 191–198). Therefore, we aimed at characterizing the adenosine receptors regulating human corpus cavernosum (HCC) smooth muscle tone. We also evaluated the pattern of extracellular catabolism of ATP in order to probe its contribution for adenosine generation in HCC. HCC specimens were collected from organ donors (control subjects) and from patients with vasculogenic erectile dysfunction (ED) according to the rules internationally accepted and approved by the Ethics Committees of HGSA and ICBAS of the University of Porto. Longitudinal strips of corpus cavernosum tissue were mounted in a 12-ml organ bath and superfused with oxygenated (95% O2 + 5%CO2) Tyrode's solution at 37°C and connected to an isometric force transducer. To test tissue relaxant responsiveness, cumulative concentration responses to adenosine and its stable analogues were evaluated in pre-contracted strips with 1 µM phenylephrine (PE). For the kinetic experiments of 30 µM ATP, AMP and adenosine catabolism in HCC strips, samples (75 µl) were collected from the bath at different times up to 45 min for HPLC analysis of the variation of substrate disappearance and product formation (Cunha & Sebastião, 1991, Eur. J. Pharmacol., 197, 83–92). We showed that HCC possesses a high ecto-NTPDase 1 activity converting ATP directly into AMP, which is then dephosphorylated into adenosine by ecto-5′-nucleotidase. Contrasting with most tissues, HCC strips slowly inactivates adenosine. Adenosine and 5′-N-ethyl-carboxamide adenosine (NECA) fully relaxed HCC (IC50 values of 30 µM and 3 µM, respectively) in control subjects. The selective A2A receptor agonist, CGS21680C, produced only a partial relaxation (30–50%) of HCC (IC50∼0.1 2M). MRS1706 (10 nM, a selective A2B receptor antagonist) attenuated NECA-induced relaxation without affecting the effect of CGS 21680C. The relaxing effects of both agonists were attenuated by ZM241385 (50 nM, a selective A2A receptor antagonist). In contrast to CGS21680C, NECA-induced relaxation was partially attenuated by NO synthase inhibition with NG-nitro-Larginine (L-NOARG, 100 µM) and by the cyclooxygenase inhibitor, indomethacin (10 µM). Patients with endothelial cell dysfunction (insensitivity to ACh) were partially resistant to NECA, but kept relaxation of HCC strips to CGS 21680C. Data indicate that extracellular adenosine causes HCC relaxation via the activation of both CGS21680C-sensitive (A2A) and -insensitive (A2B) receptors. Penile vessels from impotent men with severe vascular diseases are partially resistant to relaxation by adenosine, probably due to endothelial A2B receptor dysfunction.

Work supported by FCT (POCTI/45549/FCB/2002, participation of FEDER funding) and UP/Fundação Ilídio Pinho.

Epoxyeicosatrienoic Acids (Eets) Mediate Adenosineinduced Renal Vasodilation Via A2A Receptors

Mairéad A. Carroll, Anabel B. Doumad, Jing Li, Elvira L. Liclican and John C. McGiff

Department of Pharmacology, New York Medical College, Valhalla, New York 10562, USA Corresponding author: mairead_carroll@nymc.edu

Autocrine, endocrine and paracrine signals act on preglomerular microvessels (PGMV), where they generate a variety of mediators/modulators of the renal circulation. Activation of renal adenosine1 receptors (A1R) and A2R participate in the regulation of vascular tone and tubular function. A1R activation inhibits adenylyl cyclase (AC) via Giα, resulting in constriction of PGMV, whereas A2AR activation stimulates AC via Gsα, increasing renal blood flow. 1 The PGMV are endowed with high levels of cytochrome P450 (P450) which metabolize arachidonic acid to EETs and 20-hydroxyeicosatetraenoic acid (HETE). A dynamic and antagonistic interaction between EETs and 20-HETE is evident in their independent and opposing actions on autoregulation of blood flow; 2 20-HETE being a constrictor of PGMV and, in contrast, EETs dilate PGMV by activating calcium-dependent potassium (K+Ca2+) channels of smooth muscle cells. 20-HETE has been shown to contribute to the ATP-induced constriction of afferent arterioles via P2XR. Our study on vascular mechanisms served by EETs has uncovered a close relationship between stimulation of the A2AR and increasing P450 epoxygenase activity; 11,12-EET was identified as the likely candidate mediator of PGMV dilatation on activation of the A2AR. 3 EET levels in PGMVs increased from 1.06 to 3.86 and 7.57 ng/mg protein/15 min in response to 10 and 100 µM CGS 21680, a selective A2AR agonist, whereas, an A1R agonist failed to stimulate EET formation. Both A2AR antagonism and inhibition of epoxygenase activity prevented elevation of EET levels produced by CGS 21680.

We have investigated the commonality of the signal transduction pathway - sequential inhibition of Gsα, AC, protein kinase A (PKA) and K+Ca2+ channel activity — to the vasoactive responses to A2AR activation by CGS 21680, and 11,12-EET. Rat microdissected arcuate arteries (ca 120 µm) were cannulated and pressurized to 80mmHg. Vessels were superfused with Krebs’ solution and preconstricted with phenylephrine. 3-aminobenzamide (10 µM), an inhibitor of mono-ADP ribosyltranferases, reduced responses to 11,12-EET (3 nM) and CGS 21680 (10 µM) by 70% (p<0.05), without affecting the response to sodium nitroprusside (10 2M). Like cholera toxin (100ng/ml), 11,12-EET (100nM) stimulated ADP ribose formation in homogenates of arcuate arteries. Incubation of 11,12-EET (3nM to 3µM) with PGMV resulted in an increased cAMP levels (p<0.05), suggesting that A2AR coupled EET release is upstream of AC. An inhibitor of AC activity, SQ22536 (102M), decreased responses to CGS 21680 and 11,12-EET, reducing the internal diameter from 22±2 µm to 1±1 µm and from 24±5 µm to 9±5 µm (p<0.05), respectively. Myristolated PKI (5µM), an inhibitor of PKA, diminished the dilator responses to CGS 21680 and 11, 12-EET by 88% and 95%, respectively. The responses to both 11,12-EET and CGS 21680 were significantly reduced by superfusion of iberiotoxin (100nM), an inhibitor of K+Ca2+ channel activity. Thus, in rat PGMV, EETs mediate A2AR-induced dilation. EETs stimulate mono-ADP ribosyltransferase resulting in Gsα activation and sequential activation of AC, PKA and K+Ca2+ channel activity. We submit that EET production by PGMVs mediate the renal protective effects of adenosine. In addition, activation of the A2AR epoxygenase pathway in response to salt loading, greatly enlarges the scope of this pathway; namely, the assignment of an antihypertensive function to adenosine via release of EETs. [4]

Evaluation of adenosine A2A receptor loss in a preclinical model of Huntington disease using [11C]SCH442416 as a radioligand.

Moresco RM1,2,3, Belloli S1,2,3, Todde S1,2,3, Matarrese M1,2,3, Carpinelli A1,2,3, Turolla E1,2,3, Popoli P5, Pézzola A5, Simonelli P1,2,3, Lecchi M4.and Fazio F1,2,3.

IBFM-CNR1, Universitá degli Studi Milano Bicocca2 e Istituto Scientifico San Raffaele3, Milano, Universitá degli Studi di Milano Statale4 Milano, Istituto Superiore di Sanitá5, Roma, Italy. sara.belloli@hsr.it

Huntington's disease (HD) is a neurological disorder characterized by progressive degeneration of striatal neurons resulting in abnormal involuntary movements, psychiatric and cognitive abnormalities, and death. Adenosine A2A receptor are mainly expressed in the striatum where they are functionally linked with D2 dopamine receptors. Adenosine A2A receptor antagonists have been indicated as potential neuroprotective drugs, although the mechanisms and efficacy of these compounds remain to be elucidated. A loss of A2A receptors has been reported at 8 days after the injection of quinolinic acid (QA) a neurotoxin that produces several neurochemical modification that approximate Huntington's disease (1,2). Behavioral studies in rodents indicate that a single injection of QA induces behavioral modifications that progress over time (3). However only few studies investigated the neurochemical modifications, including the A2A receptors loss, present at later times after QA administration. Aim of this study was to evaluate whether the loss of A2A receptors observed at 8 days progresses over time and if this loss is correlated with other neurochemical modifications including dopaminergic D2 receptors loss and activation of microglial cells. For this purpose we measured ex-vivo the expression of A2A receptors until 60 days after the monolateral intrastriatal injection of QA using [11C]SCH442416 as a radioligand. Adenosine A2A receptor expression was correlated with that of D2 dopamine receptors and with the presence of activated microglial cells using respectively [11C]Raclopride and [11C]PK11195 as radioligands associated with ex-vivo or in vivo techniques. Results of the study showed a progressive decrease of [11C]SCH442416 binding in the lesioned striatum that reach a maximum reduction of approximately 58% of non lesioned striatum binding (p<0.05). A progressive loss of D2 receptors was simultaneously observed by ex-vivo and in vivo techniques (YAP-(S)PET; ISE, Italy); such a loss was maximal at 60 days after injection (60%; p<0.05). The different rate of reduction between adenosine and dopamine receptors produced a progressive reduction in D2 over A2A ratios (approximately 2.10 at 8 days until 0.8 at 60 days). Interestingly the decrease of A2A and D2 receptors was paralleled by a four times increase of [11C]PK11195 binding that was maximum at 8 days after injection (p<0.01). The results of our study indicate a progressive degeneration of intrastriatal neurons as indicated by the simultaneous loss of adenosine and dopamine receptors. Moreover, the different rate of loss of A2A and D2 receptors and the consequent modifications in their binding ratios may indicate a different response of the two neurochemical system to the neurotoxic insult. Interestingly, neuronal degeneration is followed by a constant activation of microglial cells as indicated by the increase of [11C]PK11195 binding that was maximum at 8 days and then remained stable. Finally, the results of our study confirm the feasibility of the use of molecular imaging techniques to follow disease progression in selected preclinical models of neurodegeneration.

Evidence for a patho-physiological and pharmacological role of guanine-based purines as a new extracellular signalling system.

M.P. Rathbone1 and F. Caciagli2.

Departments of Medicine1, McMaster University, Hamilton, Ontario, Canada, and Biomedical Sciences2, “G. d'Annunzio” University of Chieti-Pescara. Chieti. Italy. Email: mrathbon@mcmaster.ca

In addition to the adenine-based purinergic intercellular signaling systems, involving adenine, adenosine and adenosine phosphates, over the last 15 years analogous guanine-based systems have been discovered. Most of these are involved in “trophic” effects, affecting the growth, differentiation and survival of various cells. Indeed, guanine-based purines act synergistically with certain growth factors such as NGF, and also stimulate the production and release from cells of several growth factors and cytokines. Guanine-based purinergic signaling has been particularly investigated in cells of the nervous system and muscle. However, in addition other tissues, including skin, respond to these compounds, indicating that, like adenine based purinergic signaling, guanine based signaling may be widespread throughout many cell types and organs.

Guanine-based purines are released from cells, and when cells are damaged the release increases substantially. Indeed, under conditions simulating ischemia, cells release more guanine-based purines than adenine based purines. Moreover, the extracellular concentration of the guanine-based purines is higher than that of the adeninebased counterparts.

There is evidence that in some cases guanosine produces its effects through entering cells and interacting with an NGF-inducible protein kinase. But there is also evidence that guanosine may interact with unique receptors on the surface of cells. Similarly, there is evidence that GTP may also have cell-surface receptors that mediate some of its effects. Moreover, guanosine is metabolized to guanine by the enzyme purine nucleoside phosphorylase (PNP). Experimental studies have indicated that exogenous extracellular guanosine is relatively persistent compared to adenosine, but a large proportion of guanosine is metabolized to guanine. Emerging evidence indicates that guanine may also have its own extracellular signaling system that is distinct from guanosine. Certainly, this would be of particular interest, since the enzyme that metabolizes guanine, guanine deaminase, shows 50 fold regional variations in brain. This degree of regional variation is characteristically associated with enzymes that degrade neurotransmitters.

It appears that the concept of intercellular signaling by guanine-based purines is now well substantiated. Since GTP, guanosine and guanine have different biological effects, different receptive mechanisms and likely different signal transduction mechanisms, it could be suggested the intriguing possibility that the extracellular interconversion of these guanine derivatives provides an extra layer of signal regulation by cells.

Evidence for Extracellualr ATP in Plants: Localization and Functional Significance in Root Hair Growth and Signaling

Sung-Yong Kim1,2, Mayandi Sivaguru3, Crystal McCalvin5, and Gary Stacey1,3,4 sk2kb@mizzou.edu, staceyg@missouri.edu

1 Division of Plant Sciences, University of Missouri-Columbia. 2 National Center for Soybean Biotechnology. 3 Division of Biological Sciences and Molecular Cytology Core, University of Missouri-Columbia. 4 Division of Biochemistry, Department of Microbiogoy and Immunology. University of Missouri-Columbia. 5 Department of Microbiology, University of Tennessee-Knoxville.

Transgenic Lotus japonicus plants expressing the soybean ectoapyrase (GS52), with a predicted extracellular catalytic domain, showed increased root infection when inoculated with the nitrogen-fixing symbiotic bacterium, Mesorhizobium loti. These data suggest the presence of extracellular ATP (eATP) in plants and a role for this molecule in symbiotic infection. A variety of recent publications have shown that exogenous addition of ATP can affect plant processes. Indeed, a recent publication points to an essential role for eATP in plant cell viability. What is missing from these investigations is clear demonstration of the presence of eATP under normal plant growth conditions. We constructed a novel reporter protein where the cellulose binding domain was fused with the ATP-requiring enzyme, luciferase. This construct, in the presence of the substrate luciferin, is expected to produce luminescence only when eATP is present. Indeed, when added to legume plant roots, luminescence was seen and localized to the interstitial spaces between cells of the root and the tips of growing root hair cells. The secretion of eATP was dependent on Ca2+, as indicated by reduced luminescence upon addition of GdCl, LaCl (calcium channel blockers) or BAPTA (calcium chelator) and an increase in the intensity upon addition of CaCl2. Treatment of roots with brefeldin A, an inhibitor of vesicular trafficking, also blocked eATP release. In addition, reactive oxygen species (ROS) production is also dependent on cytosolic calcium and its gradients are also closely associated with root hair tip growth. Interestingly, addition of exogenous ATP significantly increased the production of ROS, but addition of adenosine, AMP, ADP, or non-hydrolyzable ATP (βγmeATP) did not stimulate ROS production. These data suggest that eATP is released from the root hair tip during vesicle exocytosis, concomitant with polar growth. We postulate that eATP may act on polar growth through the production of ROS. Taken together, our current work strongly suggests that eATP is present in plants and likely is an additional signal involved in the earliest events of root hair infection by rhizobia.

Evidence for the involvement of basic amino acid residues in transmembrane regions 6 and 7 of the human platelet P2Y12-receptor in ligand recognition

Kristina Hoffmann, Irina Algaier, Ivar von Kügelgen

Department of Pharmacology, University of Bonn, Reuterstrasse 2b, 53113 Bonn, Germany Kristina.hoffmann@uni-bonn.de

The P2Y12 receptor plays a crucial role in ADP-induced platelet aggregation. The ligand binding site of this receptor has yet not been fully characterized. A patient with a congenital bleeding disorder has been shown to carry a polymorphism with a change of R256 to glutamine in the upper third of transmembrane region 6 (TM6) [1]. When expressed in cells, the R256Q mutant receptor had a defective receptor function. For the P2Y1-receptor, several amino acid residues in TM6 and TM7 are involved in ligand recognition [2]. We now studied whether the corresponding amino acid residues are involved in ligand recognition of the P2Y12-receptor. Mutations were introduced in the encoding DNA sequence by site directed mutagenesis. Wild type P2Y12-receptors and mutant receptors were stably expressed in human 1321N1-astrocytoma cells. Immunofluorescence staining revealed that the expression levels of the mutant receptors were at least as high as those of wild type receptors. Receptor function was then determined by measuring the 2-methylthio-ADP-induced inhibition of cellular cAMP levels. For this purpose, cellular cAMP accumulation was increased by addition of isoprenaline 10 nM (increase of 52 ± 2 pmol cAMP per well). In cells expressing wild type receptors, 2-methylthio-ADP caused a concentration-dependent inhibition of the cAMP production with an EC50 concentration of 0.9 nM and a maximal effect of 55 % inhibition. In cells transfected with the S101A mutant receptor (TM3), the maximal response to stimulation by 2- methylthio-ADP was decreased (maximal inhibition by 29%). Next we studied mutants with a change in TM6. 2- Methylthio-ADP inhibited cellular cAMP production in cells expressing the R256K mutant receptor (change of a basic residue to another basic residue) by a similar degree as observed for the wild type receptor (EC50 concentration of about 0.2 nM, inhibition by 39 %). When R256 was replaced by alanine (R256A), there was a clear decrease in agonist potency with an EC50 concentration of about 10 nM (maximal inhibition by 23 %). In cells expressing the R256D mutant (replacement of a basic residue by an acidic residue), the maximal effect was decreased even more (only 18% inhibition left). Several mutations caused a total loss of receptor function. In cells expressing Y259D mutant receptors, 2-methylthio-ADP (0.01nM to 1 µM) failed to elicit any inhibitory effect. Similarly, 2-methylthio-ADP caused no effect in cells with the H253A/R256A double mutant receptor (TM 6). In cells expressing the K280A mutant receptor (TM7), there was also no inhibition due to stimulation by 2-methylthio-ADP. AR-C69931MX (N6-(2-methylthioethyl)-2-(3,3,3-trifluoropropylthio)-β,γ-dichloromethylene-ATP) acts as a potent antagonist at the wild type P2Y12 receptors when used in nanomolar concentrations [3]. At the R256A mutant receptor, AR-C69931MX 30 nM shifted the concentration-response curve to the right, excluding a major change in the affinity of the antagonist at this mutant receptor. In summary, our data indicate the involvement of the residues H253, R256, Y259 (TM 6) and K280 (TM 7) in ligand recognition of the human P2Y12 receptor. S101 (TM3) may play an intermediate role.

Evidence for trophic effects of guanosine in SH-SY5Y neuroblastoma cells but not in EA.hy926 and HUVEC endothelial cells

C. Florio1, S. Pacor1, U. Traversa1,2

1Department of Biomedical Sciences, 2BRAIN Centre for Neuroscience, University of Trieste, Italy. traversa@units.it

It is known that after brain injury there is early proliferation of microglia, followed by astrogliosis and then, at a later stage, proliferation of new capillary endothelial cells. The trophic effects of guanosine (GUO), which is released at high levels by hypoxia, included: i) the stimulation of the synthesis and release of neurotrophic factors which stimulate the proliferation of microglia and astrocytes and protect neurons against excitotoxic death, ii) the stimulation of PI3-K and MAPK pathways, which protect astrocytes against apoptosis. Both actions appeared to be mediated by putative GUO receptors. We questioned whether guanosine could also exert trophic effects both in endothelial and neuronal cells.

The proliferation experiments with EA.hy926 human cell line were performed in: i) different FBS conditions, starvation (0%), stand-by (1%) and proliferating (10%); ii) different exposure times (24–48 hrs) to different GUO concentrations (12.5–300µM); iii) different times of starvation (24–48–72 hrs). GUO did not significantly increase the proliferation in any conditions tested. Similarly, it was unable to modify proliferation of HUVEC cells. Studies to identify putative GUO binding sites were unsatisfactory. On the other hand, RT-PCR analysis to evaluate the expression of a orphan GPCR (G3), hypothesised to be the putative receptor for GUO since RNAsi abolished the GUO-induced phosphorylation of AKT or ERK1/2 in astrocytes, gave negative results in both cell lines. Moreover, GUO did not caused an increase in cAMP levels in the absence or presence of forskolin (3µM).

In SH-SY5Y human neuroblastoma cells a treatment for 24 h with GUO (12.5–300 µM) induced a concentrationdependent proliferative effect which was evident when the cultures were maintained in starvation (0% FBS), whereas it was not detected when the cells were cultured in 2% serum. The proliferative effect was also maintained in more severe conditions, prolonging the starvation up to 72 hrs. In this condition, when the 0% FBS medium was replaced with 2% FBS medium the cells restarted to growth and once again the addition of GUO was without effect. The cytofluorimetric analysis showed that GUO induced a concentration-dependent shift of cell cycle from G0–G1 phase towards S phase. The proliferative effects of GUO were not counteracted by the P1 receptor antagonists, DPCPX or DMPX, ruling out a role of endogenous adenosine in mediating GUO action. Suramine, a non specific P2 receptor antagonist, per se induced an increase in cell growth that was not modified by GUO. The proliferative effects of GUO were increased by the presence of 1mM propentophylline, a non specific inhibitor of purine transporters, and was significantly reduced by the pre-treatment of SH-SY5Y cells with 10µM PD098,059 and/or 30µM LY-294,002 inhibitors of MAPKs and PI-3K pathways, respectively. GUO produced a weak, but significant, increase in cAMP levels and potentiated the forskolin action which were not changed by cell treatments with ADA, DPCPX or DMPX. Preliminary experiments showed the presence of high affinity binding sites for GUO in SH-SY5Y cells that also express the G3 protein. In conclusion, these results show that guanosine exert trophic effects also on neurons with mechanisms that could be similar to those involved in its trophic action on glia, and reinforce the hypothesis of the existence of extracellular recognition binding sites for guanosine.

Exaggerated Renal Response to Adenosine in High Saltfed Rats: Role of Epoxyeicosatrienoic Acids (EETs)

Elvira L. Liclican, John C. McGiff and Mairéad A. Carroll

Department of Pharmacology, New York Medical College, Valhalla, New York 10595, USA Corresponding author: Elvira'liclican@nymc.edu

Adenosine plays a critical role in the regulation of renal vascular tone and tubular function, thus modulating renal blood flow and glomerular filtration rate. In the renal microcirculation, activation of the adenosine 1 receptor (A1R) and A2AR subtypes has opposing effects: vasoconstriction in response to A1R activation enhances proximal tubular NaCl reabsorption, whereas endothelium-dependent relaxation via A2AR promotes natriuresis1. Adenosine stimulation of A2AR in rat preglomerular microvessels is linked to production of EETs, cytochrome P-450 (CYP) epoxygenase metabolites of arachidonic acid2. EETs are important modulators of cardiovascular function and their contribution to blood pressure regulation has been established in several different animal models. EETs are thought to be natriuretic by virtue of their ability to dilate the renal vasculature as well as regulate Na+ transport in proximal and distal tubules. An increase in natriuretic EETs is one of the significant components of the kidney's adaptive response to prevent elevation of blood pressure in response to high salt (HS) intake. It is also documented that renal ARs are affected by Na+ intake3, and mice lacking A2AR exhibit elevated blood pressure4. As adenosine levels are increased by dietary salt intake, we propose that adenosine is the stimulus for increased renal epoxygenase activity in response to salt loading. More specifically, we hypothesize that HS intake increases the renal response to adenosine, resulting in increased epoxygenase activity and EET levels.

Male Sprague-Dawley rats were fed HS (4.0% NaCl) or normal salt (NS; 0.4% NaCl) diet. On day 8, isolated kidneys were perfused with Krebs’ buffer and preconstricted to ca 150 mm Hg with phenylephrine (10−7M). Renal effluent eicosanoids were analyzed by GC/MS. Bolus injections of the stable adenosine analog 2-chloroadenosine (2-CA; 0.1-10µg) resulted in dose-dependent dilation; at 10µg, perfusion pressure (PP) was lowered to a greater extent in the kidneys of HS rats compared to NS rats (−60 ± 4 vs. −31 ± 8 mm Hg; p<0.05) and the area of response was increased (27 ± 6 vs. 9 ± 4 mm2; p<0.05), as was EET release (132 ± 23 vs. 38 ± 18 ng; p<0.05). HS treatment increased A2AR and CYP2C23 epoxygenase protein expression. A selective epoxygenase inhibitor, N-methylsulfonyl-6-(2-propargyloxyphenyl) hexanamide (MS-PPOH; 12 µM), significantly reduced the response to 2-CA in HS rats; PP, area of response, and EET release decreased by 40%, 70% and 81%, respectively, whereas lesser changes were evident in NS kidneys. In addition to the increase in EET levels with HS intake, a ca 97% increase of total purines (as an index of flux through the adenosine pathway) was measured in renal effluents obtained from HS-fed rats compared to NS-fed rats (2357 ± 268 vs. 1197 ± 127 nmol/L; p<0.05). In vivo, infusion of MSPPOH (5mg/kg daily for 3 days) via osmotic pumps increased blood pressure and decreased UNaV in HS-compared to NS-fed rats.

Thus, the greater vasodilator response to 2-CA seen in kidneys of HS-fed rats was mediated by increased EET release, presumably via upregulation of A2AR and epoxygenase expression. As EETs have renal vasodilator and natriuretic properties, A2AR activation may be a key link in a renal mechanism that contributes to the regulation of blood pressure.

“Exchange Protein Activated by cAMP” (Epac)-mediated “Suppressor of Cytokine Signalling-3” (SOCS-3) induction: a novel A2A adenosine receptor-triggered anti-inflammatory signalling pathway

William A. Sands, Hayley Woolson, Gillian R. Milne and Timothy M. Palmer

Division of Biochemistry & Molecular Biology, IBLS, University of Glasgow, Scotland, U.K. Presenting author e-mail: W.Sands@bio.gla.ac.uk

While the anti-inflammatory effects of the A2A adenosine receptor (A2AAR) are well-established, the molecular mechanisms responsible for these effects remain unclear. Here we demonstrate that selective activation of endogenous A2AARs by CGS21680 induces the expression of “suppressor of cytokine signalling-3” (SOCS-3) but not the related protein SOCS-1 in human umbilical vein endothelial cells (HUVECs). The effect could be recapitulated by elevation of intracellular cyclic AMP (cAMP) but was resistant to inhibition of PKA and was instead mimicked by either selective activation of the cAMP-activated Rap GEF “Epac” or transient expression of constitutively active Val12Rap1a. Importantly, SOCS-3 induction was associated with a PKA-independent ability of cAMP to inhibit IL-6-mediated STAT1 and STAT3 activation by an interleukin-6 (IL-6) trans-signalling complex comprising IL-6 and soluble IL-6 receptor-α (hereafter termed IL-6/sIL-6Rα) via gp130, a well-characterised target for inhibition by SOCS-3. Attenuation of SOCS-3 induction using either morpholino antisense-mediated knockdown or employing SOCS-3-null murine embryonic fibroblasts (MEFs) abolished the inhibitory effect of cAMP on IL-6/sIL-6Rα-activated STAT phosphorylation whereas inhibition of SHP-2, another key negative regulator of gp130, was without effect. These data suggest that SOCS-3 is the dominant mediator of cAMP's effects. Interestingly, cAMP elevation activated the ERK pathway in HUVECs, and this was required to observe SOCS-3 induction. Subsequent in silico analysis of the human SOCS-3 promoter identified several potential sites for ERK regulation, including consensus motifs for binding of the CAAT enhancer binding protein (CEBP) transcription factor family. Chromatin immunoprecipitation assays confirmed that either elevation of cAMP or selective activation of Epac stimulated the rapid accumulation of CEBPβ on the SOCS-3 promoter in HUVECs. Together, these data argue for the existence of a novel cAMP/Epac/CEBPβ/SOCS-3 pathway for limiting pro-inflammatory signalling from Class I cytokine receptors in endothelial cells, and illuminate a new mechanism by which the A2AAR may mediate its potent anti-inflammatory effects.

Exposure to Hg2+ and Pb2+ alters NTPDase and ecto-5′-nucleotidase activities in central nervous system of zebrafish (Danio rerio)

Carla Denise Bonan1, Mario Roberto Senger 1,2, Eduardo Pacheco Rico 1,2, Marcelo de Bem Arizi1, Ana Paula Guedes Frazzon3, Renato Dutra Dias1, Maurício Reis Bogo1.

1Faculdade de Biociências, Pontifícia Universidade Cató lica do Rio Grande do Sul. Porto Alegre, RS, Brazil 2Departamento de Bioquímica, UFRGS, Porto Alegre, RS, Brazil. 3Departamento de Farmacologia e Toxicologia, FFFCMPA, Porto Alegre, RS, Brazil. cbonan@pucrs.br

Neurotransmission can be affected by exposure to heavy metals, such as mercury and lead. ATP is a signaling molecule that can be inactivated by ecto-nucleotidases. Ecto-nucleotidases are ubiquitous enzymes with a broad phylogenetic distribution, occurring in many vertebrate tissues. Zebrafish is a consolidated model system in neuroscience and toxicological studies. The zebrafish genome project has demonstrated regions of syntenic relationship with human genome. Purinoceptors were already identified in this teleost and, recently, we characterized the presence of a NTPDase and an ecto-5′-nucleotidase activities in brain membranes of zebrafish. Considering that mercury and lead are important environmental contaminants and the presence of purinergic receptors and enzyme activities involved in extracellular catabolism of nucleotides in zebrafish brain, the aim of present study was to investigate the effect of mercury chloride (HgCli2) and lead acetate [Pb(CH3COO)2] on NTPDase and ecto-5′-nucleotidase activities and expression in central nervous system of zebrafish. In vitro exposure to HgCl2 decreased ATP and ADP hydrolysis in an uncompetitive mechanism and AMP hydrolysis in a non-competitive manner in brain membranes of zebrafish. Pb(CH3COO)2 inhibited ATP hydrolysis in an uncompetitive manner, but not ADP and AMP hydrolysis. In vivo exposure to HgCl2 or Pb(CH3COO)2 (20 µg/L, during 24h, 96h and 30 days) caused differential effects on ecto-nucleotidase activities. Exposure to HgCl2 during 96 h caused a significant inhibition of ATP (37%), ADP (42%) and AMP (39%) hydrolysis Interestingly, after 30 days of exposure to HgCl2, ATP hydrolysis return to the control level and ADP hydrolysis was strongly increased (118%) when compared to the control values. AMP hydrolysis at this time of exposure remaining inhibited (32%). However, after 96 h of exposure to Pb(CH3COO)2, it is possible to observe a significant decrease on ATP hydrolysis (17%), but not on ADP and AMP hydrolysis. The chronic exposure to Pb(CH3COO)2 during 30 days also promoted a significant decrease of ATP (33%), ADP (37%) and AMP (40%) hydrolysis in brain membranes of zebrafish. To verify if the chronic exposure to mercury or lead during 30 days was able to modify ecto-nucleotidases expression, RT-PCR experiments were performed. There were no changes in the expression of NTPDase1 and 5′-nucleotidase, following 30 days of exposure to both metals. The results indicate that the significant alterations observed probably are related to post-translational changes of the enzymes. Therefore, this study demonstrated that ectonucleotidases can be a potential target related to neurotoxicity induced by mercury and lead and a possible indicator of the biological impact of exposure to heavy metal contaminants.

Supported by: FAPERGS, CNPq, CAPES, TWAS.

Expression and function of A2A and A3 adenosine receptors in mouse neutrophils

Dharini van der Hoeven, John A. Auchampach

Department of Pharmacology and Toxicology, Medical College of Wisconsin, WI, USA jauchamp@mcw.edu

Adenosine is a potent anti-inflammatory agent, which partially explains its beneficial effects when administered in models of tissue injury and inflammation including ischemia/reperfusion injury. The aims of this study were to comprehensively characterize the expression profile of the four AR subtypes in mouse neutrophils and to determine their roles in regulating neutrophil functions. Neutrophils were isolated from bone marrow by Percoll density gradient centrifugation and immunomagnetic selection with an anti-Gr-1 antibody resulting in isolates that were 99% pure based on cytospin images and FACS analysis. Real-time PCR analysis of purified neutrophil populations revealed that mRNA expression levels of the A2A and A3AR were highest (3809.32 ± 391.38 and 3524.19 ± 685.17 copies of mRNA/50 ng of total RNA, respectively) followed by the A2BAR (162.89 ± 32.87 copies of mRNA/ 50 ng of total RNA). mRNA expression of the A1AR was negligible. Radioligand binding analysis using the A2AAR antagonist 125I-ZM-241385 and the A1/A3AR agonist [125I]I-AB-MECA also suggested that A2A and A3AR proteins are abundantly expressed in mouse neutrophils, confirming the results of the quantitative PCR studies. We subsequently examined the role of A2A and A3ARs in regulating neutrophil superoxide (O2.−) production and chemotaxis. In O2.− assays, Percoll gradient purified neutrophils were treated with AR agonists for 30 minutes prior to stimulation with activating agents and O2.− produced was measured using the chemiluminescent probe 2-methyl-6-(p-methoxyphenyl)-3,7-dihydroimidazo [1,2-alpha]pyrazin-3-one (MCLA, 0.5 µM). The non-specific AR agonist NECA (300 nM), the A2AAR agonist CGS-21680 (100 nM) and the A3AR agonist CP-532,903 (100 nM) inhibited O2.−production by wild-type (WT) neutrophils activated with N-formyl peptide fMLP (1 µM), complement C5a (3 nM) and platelet activating factor (PAF, 100 nM), but not with phorbol myristate acetate (PMA, 800 nM). Subsequent concentration-response studies revealed that both CGS-21680 and CP-532,903 inhibited O2.− production by neutrophils with similar potencies (EC50 values of 16.43 ± 2.07 nM and 38.10 ± 1.60 nM respectively) and efficacies (∼45% and ∼55% inhibition, respectively). In chemotaxis assays, Percoll gradient purified WT neutrophils were treated with CGS-21680 or CP-532,903 (100 nM) for 30 minutes and directed migration towards fMLP (30 nM–10 µM) or C5a (300 pM to 100 nM) was measured using a standard 48-well chemotaxis chamber with polycarbonate membranes (5 µ pore size). Activation of the A3AR with 100 nM CP-532,903 (but not the A2AAR with CGS 21680) inhibited fMLP-induced chemotaxis of mouse neutrophils producing ∼25% inhibition elicited by 1 and 3 µM fMLP. Interestingly, CP-532,903 did not inhibit C5a-induced neutrophil chemotaxis. In both the assays for O2.−production and chemotaxis, the specificity of the agonists for A2A and A3ARs was confirmed by repeating the experiments with neutrophils obtained from A2AAR or A3AR gene ‘knock-out’ mice. In conclusion, while activation of the A2A or A3AR inhibits neutrophil O2.− production, activation of the A3AR may also lead to stimulant-specific inhibition of neutrophil chemotaxis.

Expression and Function of Purinergic Receptors in Human Type-B Synoviocytes. A Preliminary Study

1F. Caporali, 2A. Gamberucci, 1G. Pompella, 1P.L. Capecchi, and 1F. Laghi Pasini.

1Department of Clinical Medicine and Immunological Science, Division of Clinical Immunology, and 2Department of Pathophysiology, Experimental Medicine, and Public Health, University of Siena, Siena, Italy. caporali7@unisi.it

Type-B synoviocytes are actively involved in joint injury during chronic and acute rheumatic diseases producing cytokines and inflammatory mediators.

The aim of this study was to investigate P2 receptors (P2R) expression and role in human B-type synoviocytes obtained from articular biopsies.

mRNA levels of P2R were analysed by RT-PCR. Human primary synoviocytes were found to express mRNA for the following P2X and P2Y receptors: P2X1, P2X2, P2X4, P2X5, P2X6, P2X7, and P2Y1, PY4, P2Y11, P2Y12, P2Y13 and P2Y14. Moreover we observed P2X7 expression by Western Blotting.

We analyzed whether synovial cells have the ability to mobilize calcium when challenged with agonists that interact with the purinergic receptors family.

ATP (1mM) induced an increase in intracellular free calcium concentration ([Ca2+]i) in FURA-2 loaded synoviocytes in presence and in absence of extracellular calcium, suggesting that calcium rise depends either on influx across the plasma membrane or on release from intracellular stores and so both P2X and P2Y receptors appear to be involved in calcium mobilization in synoviocytes.

To determine the specific role of P2X receptors we stimulated synoviocytes with a P2X7 receptor agonist, BzATP (500µM),which was able to induce a rise in [Ca2+]i in the presence of extracellular calcium; conversely, no changes in [Ca2+]i were observed when BzATP was tested in the absence of extracellular calcium.

Since BzATP was reported to activate also P2X1 and P2X3 receptors, we investigated the calcium mobilizing activity of α,βMeATP, a specific agonist of these receptors. Indeed after α,βMeATP (100 µM) stimulation no significant [Ca2+]i rise was observed, indicating the possible P2X7 receptor activation.

On the basis of the previous considerations we investigated the involvement of P2X7 receptor in the IL-6 release in ELISA experiments through BzATP stimulation.

Synoviocytes spontaneously released low levels of the proinflammatory cytokine interleukin-6 (IL-6) as measured in the supernatant by ELISA. Stimulation with IL-1β increased such secretion after 3–6–24h incubation. We observed a time-dependency in the secretion of IL-6 after BzATP (100–500µM) treatment. In agrement with the hypothesis about the P2X7 receptor involvement we used a P2X7-selective antagonist, oxidized ATP and we observed that the increase in the cytokine release was inhibited.

This very preliminary study suggests that: i) human synoviocytes express mRNA for several P2X and P2Y receptors; ii) P2X7 activation is probably associated with cytokine release; iii) P2R antagonism may represent a therapeutic target for the pharmacological manipulation of the proinflammatory activity of synoviocytes.

Expression and role of the adenosine A2a receptor in the dorsal horn of rat spinal cord

H. Dumont1,2, E. Guntz1,2, D. Gall2, A. de Kerchove d'Exaerde2, S.N. Schiffmann2, M. Sosnowski1.

1 CHU St Pierre, Brussels, Belgium 2 Lab. Neurophysiology, ULB, Brussels, Belgium maurice_sosnowski@stpierre-bru.be

Background and goal of the study

Adenosine is an endogenous neuromodulator that acts on 4 different subtypes of G-coupled protein receptors (A1, A2a, A2b, A3). The presence of A2a receptor is well established in the central nervous system, predominantly in the striatum where A2a receptor activation was described as a modulator of the NMDA receptor (1,2). In contrast, the distribution and the role of A2a receptor in the spinal cord remain unclear (3). The goal of this study is to explore the presence and the role of A2a receptor in the spinal cord pain pathways.1,2). In contrast, the distribution and the role of A2a receptor in the spinal cord remain unclear (3). The goal of this study is to explore the presence and the role of A2a receptor in the spinal cord pain pathways.

Material and Methods

Total RT-PCR was first performed to assess the expression of the adenosine A2a receptor gene in the lumbar enlargement of the rat spinal cord. Second, single-cell RT-PCR was performed on acute slices in order to identify cells expressing A2a receptor mRNA and to quantify the proportion of lumbar dorsal horn neurons that express it. Third, we investigated whether the activation of A2a receptor has any modulatory effect on the activity of NMDA receptor by using the whole cell patch clamp technique on projection neurons from lamina II.

Results and Discussion

RT-PCR performed on the entire lumbar spinal cord revealed the presence of the adenosine A2a receptor transcript. RT-PCR performed on single cell identified as projection neurons revealed the presence of the adenosine A2a receptor transcript in 5 out of 32 cells. Electrophysiological recordings did not show a significant difference between the current induced in presence of 10 µM NMDA and the one induced in presence of 10µM NMDA plus 0.1µM CGS 21680, the A2a receptor agonist (−107.32 ± 16.93 pA vs −83.43 ± 9.31 pA, n=7, student's t-test, p>0.05).


Adenosine A2a receptor is present in the dorsal horn of the lumbar spinal cord. The receptor is expressed on 15 % of the lamina II projection neurons. A2a receptor agonists seem not able to modulate the NMDA receptor activity.

Expression of ecto-alkaline phosphatase in the murine neurogenic subventricular zone

D. Langer1, Y. Ikehara2, H. Takebayashi3 and H. Zimmermann1

1Institute of Cell Biology and Neuroscience, Biocenter, J.W. Goethe-University, 60439 Frankfurt, Germany, 2Department of Cell Biology, Fukuoka University School of Medicine, Fukuoka 814-0180, Japan, 3Division of Molecular Neurobiology and Bioinformatics, National Institute for Physiological Sciences, Okazaki 444-8787, Japan D.Langer@zoology.uni-frankfurt.de

In the adult murine brain, neurogenesis persists in two brain regions, the subgranular layer of the hippocampal dentate gyrus and the subventricular zone (SVZ) of the lateral ventricles. In the SVZ, multipotential and selfrenewing astrocyte-like stem cells (type B cells) are the source of newly generated neuroblasts (type A cells). The highly proliferating type C cells are thought to represent intermediates between type A and type B cells. The neuroblasts migrate along the ventricular surface and the rostral migratory stream (RMS) towards the olfactory bulb (OB). Forming a network of tightly associated cells, the migrating cells pass through tunnels formed by processes of type B cells. In the OB, the type A cells move radially and differentiate into granular or periglomerular interneurons. We have previously shown that the ecto-nucleotidase NTPDase2 is selectively associated with type B cells of the SVZ1 and neural progenitors in the hippocampal dentate gyrus2 and that neurosphere cells cultured from SVZ stem cells express functional P2Y1, P2Y2 and P1 receptors whose activation synergistically supports growth factor-mediated cell proliferation.3

Using enzyme histochemistry and immunocytochemistry we demonstrate that an additional ecto-nucleotidase is expressed by cells of the adult murine SVZ and RMS, the tissue non specific form of alkaline phosphatase (TNAP). TNAP is capable of hydrolyzing nucleoside tri-, di- and monophosphates with an alkaline pH optimum. In contrast to NTPDase2, TNAP can generate extracellular adenosine. Interestingly, TNAP is not expressed by adult hippocampal progenitors. Using double immunohistochemistry and various cell markers, we demonstrate that TNAP does not colocalize with the NTPDase2-expressing type B cells. It rather is associated with their descendants, the Olig2-expressing type C cells and the doublecortin or PSA-NCAM-expressing type A cells. At physilogical pH, the ATP-hydrolyzing activity of TNAP in the neurogenic zone is lower than that of NTPDase2. During brain ontogeny, TNAP is already expressed within the mouse embryonic neurogenic zones at day 14 (E14), several days before NTPDase2. Taken together these results further support the notion that signaling via extracellular nucleotides and presumably also nucleosides supports both embryonic and adult mammalian neurogenesis. Ecto-nucleotidases would have the potential to control the availability of agonists at P2 and P1 receptors whose cellular expression pattern in the neurogenic zones requires further investigation.

Expression of Human Ecto 5T'-Nucleotidase in Pig Endothelium and its Effects on Adenosine Production, NK Cell-Mediated Lysis and Platelet Function

Foy N Osborne 1), Kameljit K Kalsi 1), Charlotte Lawson 1), Marialuisa Lavitrano 2), Magdi H Yacoub 1), Marlene L Rose 1), Ryszard T Smolenski1)

1) Heart Science Centre, Imperial College at Harefield Hospital, Harefield, U.K.

2) Department Medicina Sperimentale Ambientale e Biotecnologie Mediche, University of Milano-Bicocca, Via Cadore, 48, 20052 Monza, Milano, Italy r.smolenski@ic.ac.uk

The use of xenogeneic cells, tissues and organs is one possible solution to circumvent the shortage of human organs for allotransplantation. Pigs are considered as optimal candidates and major obstacle after transplantation of pig organs into primates, hyperacute rejection has been partially resolved in transgenic pigs expressing human complement regulatory proteins such as human decay accelerating factor (hDAF) or by knocking-out α1,3-galactosyltransferase. However, delayed xenograft rejection or acute vascular rejection that occurs within days to weeks after pig-to-primate organ transplantation remains unresolved. This process is mediated by immune and haemostatic mechanisms involving antibodies, natural killer (NK) cells and monocytes. Ecto-5′-nucleotidase (E-5′N) is an endothelial surface enzyme that controls conversion of extracellular nucleotides that triggers thrombosis and immune response into into immunosuppressive and antithrombotic adenosine. We evaluated whether expression of human E50N on pig endothelial cells (EC) attenuates human NK cell mediated cytotoxicity and inhibit platelet aggreggation and adhesion.

A pig EC line was stably transfected with human E5′N and human NK cell adhesion and cytotoxicity towards pig EC cultures was measured by flow cytometry and intracellular enzyme release. E5′N activity in pig EC lysates increased from 0.68 ± 0.07 to 1013 ± 293nmol/min/mg protein, whilst the rate of AMP to adenosine metabolism by intact cells increased from 0.37 ± 0.05 to >300nmol/min/mg protein in non-transfected and transfected cells, respectively. The rate of adenosine production in transfected cells increased also with ATP as the extracellular substrate. Cytotoxicity of human NK cells was reduced from 10.7 ± 0.4% and 11.1 ± 1.1% with non-transfected pig EC to 5.2 ± 0.2% and 5.0 ± 0.2% in transfected cells with 50µM and 250µM AMP respectively. Reduction of cytotoxicity in E5′N-transfected EC was abolished by the E5′N inhibitor and was mimicked in non-transfected EC by the addition of adenosine, demonstrating the key role of adenosine produced by E5′N in inhibiting NK cell cytotoxicity. Platelet aggregation measured using aggregometer was markedly reduced by supernatants of cells transfected with E5′N added to platelet reach plasma. Adhesion of fluorescently labelled platelets to E5′N transfected pig endothelial cells was reduced to 39 ± 9% of the adhesion observed in non-transfected cells. Both effects on platelet function were abolished by the E5′N inhibitor. We suggest that over-expression of E5′N in EC of transgenic pigs is a possible strategy to ameliorate rejection after xenotransplantation.

Extracellular ATP: a critical modulator of hypoxia-induced pulmonary artery adventitial fibroblast proliferation

E.V. Gerasimovskaya1*, D.A. Tucker1, M. Weiser-Evans2, J.M. Wenzlau3, S. Ahmad4, C.W. White4, K.R. Stenmark1,

1Department of Pediatrics and 2Department of Medicine University of Colorado at Denver and Health Sciences Center, Barbara Davis Center for Childhood Diabetes and JMRC3, Denver, CO, USA; *E-mail: Evgenia.Gerasimovskaya@UCHSC.edu

Nucleotides such as ATP and UTP are emerging as a ubiquitous family of extracellular signaling molecules. They have been shown to play a key role in transducing mitogenic, contractile, metabolic, and secretory signals in a variety of cells through release and subsequent binding to the P2 (P2X and P2Y) family of purinergic receptors1,2. However, little is known regarding an autocrine role for ATP in hypoxia-induced vascular cell responses. In a neonatal model of hypoxic pulmonary hypertension, we showed that Pulmonary Artery (PA) adventitial fibroblasts proliferate both in vivo and in vitro in response to hypoxic conditions3,4. We therefore examined the hypothesis that hypoxia-induced adventitial fibroblast proliferation would be mediated by hypoxia-induced changes in ATP release and/or its extracellular degradation. We found that acute hypoxia (3%O2 10–60 min) increased extracellular ATP concentrations in adventitial fibroblasts and that chronic hypoxia (3%O2, 14–30 days) markedly attenuated the rate of extracellular ATP hydrolysis by ectonucleotidase(s) suggesting that at least two different cellular mechanisms may contribute to elevated extracellular ATP levels. Exogenous ATP (100 µM) stimulated thymidine incorporation and increased the phosphorylation of Akt, Erk1/2, mTOR, and p70S6K in fibroblasts as did UTP, UDP, ADP, ADPβS, MeSATP, αβMeATP, BzATP and some other agonists, indicating that both P2Y and P2X purinoceptors mediate mitogenic responses. PCR analysis revealed that adventitial fibroblasts express P2Y1,2,6 as well as P2X2,4,6,7 receptor subtypes. The rank order of potency of various agonists to activate each individual kinase pathway (ERK1/2, PI3K/Akt or mTOR/p70S6K) indicates that in these cells, purinergic receptors are coupled to the proliferative responses in a pathway-specific manner. Importantly, in this ATP-activated signaling network, a translational pathway, involving mTOR, p70S6K and S6 ribosomal protein, plays a central role in integrating Erk1/2 and PI3K/Akt pathways. We also found that ATP (100µM) and hypoxia (3%O2), induced expression and activation of the Egr-1 transcription factor, and both stimuli acted, in part, through Gαi-initiated ERK1/2 signaling pathway. Apyrase (2.5U/ml), as well as the non-selective P2 receptor antagonists, suramin, cibacron blue 3GA, and PPADS (all used at 100µM) attenuated hypoxia- and ATP-induced DNA synthesis, indicating an activation and a functional role of P2Y/P2X purinoceptors in hypoxia-induced proliferative responses. In addition, suramin, cibacron blue 3GA and apyrase, markedly attenuated hypoxiainduced ERK1/2 activation and Egr-1 expression. Collectively, our findings demonstrate that PA adventitial fibroblasts can be considered as endogenous source and a target of extracellular nucleotides within the vascular wall and that a hypoxia-induced autocrine loop of ATP signaling plays a critical role in the regulation of fibroblast proliferation under hypoxic conditions.

Extracellular ATP: a potential regulator of vasa vasorum neovascularization in hypoxia-induced pulmonary vascular remodeling

E.V. Gerasimovskaya1*, S. Riddle1, D.A. Tucker1, S. Ahmad2, C.W. White2, and K.R. Stenmark1

1 Department of Pediatrics University of Colorado at Denver and Health Sciences Center; and 2National Jewish Medical and Research Center, Denver, CO *E-mail: Evgenia.Gerasimovskaya@UCHSC.edu

Pathological vascular remodeling is a key component and frequently life-threatening consequence, of vascular diseases in both the systemic and pulmonary circulation1,2. In a neonatal model of hypoxic pulmonary hypertension, we have demonstrated that hypoxia-induced pulmonary artery (PA) remodeling is associated with marked increases in adventitial thickening and the expansion of vasa vasorum network2. However, the precise cellular and molecular mechanisms contributing to these changes remain unclear. Extracellular adenine nucleotides are increasingly recognized as important regulators of vascular function3,4}. Since hypoxia has been shown to stimulate ATP release in endothelial cells and fibroblasts5,6, we hypothesized that the endothelium of newly forming adventitial vasa vasorum may represent another, as yet unidentified, source of extracellular ATP and that ATP may mediate or modulate angiogenic responses in vasa vasorum endothelial cells (VVEC). To test this hypothesis, VVEC were isolated from PA adventitia of chronically hypoxic calves and characterized for the expression of endothelial markers. We found that acute hypoxic exposure (1% and 3% O2, 10–60 min) induced an increase in extracellular ATP in VVEC conditioned media from 0.02 to 2.12 nmol. Quinacrine staining revealed ATP containing vesicles in the cytosol and around the nuclei. We showed that hypoxia-induced ATP release could be attenuated by the PI3K inhibitors LY294002 (20 µM) and Wortmannin (1µM); the Rho-kinase inhibitor Y27632 (10 µM); and by the inhibitor of exocytosis NEM (4 µM) suggesting that regulated exocytosis and PI3K/Rho kinase pathways are involved in hypoxia-stimulated ATP release. We also found that exogenous ATP and hypoxia stimulated growth and tube formation in VVEC and induced activation of PI3K/Akt, ERK1/2, and mTOR/p70S6K, all of which are known to be associated with mitogenic and angiogenic signaling in endothelial cells. Notably, the magnitude of ATP-induced mitogenic responses in VVEC reached 20 fold, suggesting that these cells represent physiologically relevant sources and the targets of adenine nucleotides. Finally, we demonstrated that hypoxia-induced mitogenic responses in VVEC were significantly attenuated by the P2 receptor antagonists, suramin (100 µM) and cibacron blue 3GA (100 µM). In summary, our data support the idea that ATP released by VVEC into the local adventitial environment may serve as a potent autocrine/paracrine factor contributing to vasa vasorum neovascularization in hypoxic pulmonary hypertension. The characterization of purinergic receptors and coupled signaling pathways in VVEC will provide unique targets for therapeutic strategies aimed at inhibiting pathologic angiogenesis in the blood vessel wall.

Extracellular ATP acting on murine RAW 246.7 macrophages enhances thrombin generation in the absence of cell death

Moore, S.F. & MacKenzie, A.B.

Department of Pharmacy and Pharmacology, University of Bath, Bath. BA2 7AY. U.K. sfm22@bath.ac.uk

Extracellular adenosine 5′triphosphate (ATP) acting at the P2X7 receptor (P2X7R) has been shown to be important for macrophage inflammatory responses including the release of pro-inflammatory cytokines and formation of reactive oxygen species. Host inflammatory responses are closely linked to coagulation where macrophages can participate in events leading to the formation of thrombin in atherosclerosis and inflammation1. Cloned P2X7Rs couple to the rapid externalization of phosphatidylserine (PS), a cofactor for the assembly of the prothrombinase complex, in the absence of cell death2. The objective of the present study was to characterize the pharmacological properties of endogenous murine P2X receptors expressed by a RAW264.7 macrophage cell line and the potential role in prothrominase activity.

Fluorescent measurements of intracellular Ca2+ concentrations in Fluo-4-AM loaded cells or Ethidium Bromide (EtBr, MW 390 Da) influx have been used to evaluate functional P2X receptor expression. Unless stated, all experiments were performed in saline containing 147 mM NaCl, 2 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 12 mM glucose and 10mM HEPES (pH 7.4) at 37 °C. ATP (0.1 µm — 3 mM) elicited a dose dependent increase in intracellular Ca2+ with a peak response at 3 mM. ATP-stimulated a dose-dependent increase in EtBr uptake with an EC50 944 ± 7.4 µM (n= 3) providing evidence for the expression of the P2X7R. Using 1mM ATP, EtBr uptake was observed to increase 2-fold with the removal of extracellular divalent cations (n=3) but was unaffected by replacement of extracellular Na+ with 154 mM N-methyl D-glucamine (n=3) or isotonic KCl (n=3). Pre-application of 3 µM Ivermectin did not alter ATP induced EtBr uptake (n=3). EtBr uptake evoked by 3mM ATP was blocked by KN-62 with an IC50 1.25 ± 0.001 µM (n=3).

Brief activation of P2X7Rs (<10 mins) is associated with PS externalization3 and reversible membrane blebbing. Exposure of PS on the outer leaflet of the plasma membrane is associated with a number of macrophage functions such as the adhesion to endothelium4 and the assembly of the procoagulant enzyme complexes1. Prothrombin is activated to thrombin by the enzyme complex termed “prothrombinase” consisting the enzyme factor Xa, cofactor/factor Va, Ca2+ and negatively changed phospholipids most commonly PS. Assays performed in the presence of factors Va and Xa demonstrate that exposure to 3 mM ATP (>2 mins) enhanced the prothrombinase activity (3 fold) on macrophages. By contrast, only prolonged stimulation with 3 mM ATP (>30 mins) triggered the release of lactate dehydrogenase which was blocked by KN-62 (IC50 = 0.283 ± 0.04 µM, n=3). This data indicates that extracellular ATP can rapidly trigger an increase in prothrombinase activity in the absence of cell death.

Extracellular ATP and adenosine — two important immunomodulators regulating leukocyte-endothelial adhesion

Gennady G. Yegutkin*1, Tiina Henttinen2, Andrey Mikhailov2, Sergei Samburski1, Sirpa Jalkanen1

1 MediCity Laboratory and Department of Medical Microbiology, Turku University and National Public Health Institute, FIN-20520 Turku, Finland 2 Turku Centre for Biotechnology, University of Turku/Åbo Akademi University, POB 123, 20521 Turku, Finland * E-mail: gennady.yegutkin@utu.fi

Leukocyte-endothelial interactions involve both cell adhesion through specific receptor-ligand pairs and bidirectional cell signaling, and are affected by soluble mediators, which modulate adhesive molecules and signaling events in both cell types. Extracellular ATP and ADP are important signaling molecules mediating leukocytosis, thrombosis and several inflammatory responses in the vasculature, whereas adenosine has a non-redundant counteracting role in the attenuation of inflammation and tissue damage in vivo. The duration and magnitude of purinergic signaling is governed by a network of ectoenzymes and endothelial and lymphoid cells are generally characterized by counteracting, ATP-inactivating and ATP-regenerating/adenosine-eliminating, phenotypes, respectively. Here we have identified a link between the adhesion cascade and extracellular purine turnover. Upon adhesion, lymphocytes prevent adenosine formation in the endothelial microenvironment that, as a consequence, impairs the vascular barrier function and facilitates subsequent step of leukocyte transmigration into the tissue. Together, these leukocyte adhesion-mediated shifts in the local nucleotide and nucleoside concentrations may represent a previously unrecognised paracrine mechanism affecting the functional state of the targeted endothelium and co-ordinately regulating lymphocyte trafficking between the blood and tissues.

Extracellular ATP inhibits neutrophil apoptosis by increasing intracellular cAMP, but independently of increases in intracellular calcium levels, via P2-purinoceptors

Kathryn R. Vaughan1, Leanne Stokes2, Annmarie Surprenant2, Moira K. Whyte1.

1Academic Unit of Respiratory Medicine, School of Medicine and 2Biomedical Sciences, University of Sheffield, UK K.R.Vaughan@sheffield.ac.uk

At sites of inflammation ATP is released from activated platelets and endothelial cells, and delays neutrophil apoptosis in a concentration- and time-dependent manner, with significant effects occurring within 10 minutes of ATP exposure (% apoptosis at 5 hours: 23.0 ± 4.9 (mean T SEM) control vs 9.9 ± 1.2 10µM ATP, p<0.05). Typical P2X-receptor agonists BzATP and βγMeATP, the stable ATP analogue ATP-γ-S, and the typical P2Y-receptor agonist 2MeSATP, delayed neutrophil apoptosis to levels comparable to ATP, whereas the P2Y-agonist UTP was without effect. We found highly-pure (i.e. PBMC depleted) populations of human neutrophils express P2X-1, P2X-5, P2Y-1, P2Y-2, P2Y-4, P2Y-6 and P2Y-11 receptor mRNA. Increases in intracellular calcium [Ca2+]i were demonstrated by Ca2+ fluorimetry to be induced by UTP and thus independent of the delay of neutrophil apoptosis. We then investigated other possible mechanisms by which ATP inhibits apoptosis. Inhibition of cAMP-dependent protein kinases completely abrogated ATP-mediated neutrophil survival (% apoptosis at 16 hours: 46.2 ± 5.0 control vs 27.5 ± 3.9 100µM ATP, p<0.05, vs 56.4 ± 6.3 0.7mM Rp-8-Br-cAMPS with 100µM ATP, p<0.01), whilst ATP increased intracellular cAMP [cAMP]i levels. These data suggest that ATP-mediated delays in neutrophil apoptosis are likely P2Y-11 receptor mediated via increases in [cAMP]i, and that these effects are not associated with increases in [Ca2+]i.

Extracellular ATP signaling affects contractility of skeletal muscle

D. Sandonà1, D. Danieli-Betto 2, E. Germinario 2, S. Gastaldello 1, R. Betto3

1 Department of Biomedical and Experimental Sciences, University of Padova, 2 Department of Human Anatomy and Physiology, University of Padova, and 3 Laboratory of Muscle Biology and Physiopathology, C.N.R. Institute of Neuroscience, Italy. (romeo.betto@bio.unipd.it)

Muscle contraction consumes more than 50% of the ATP produced by the striated cells, moreover, the working muscle also releases some of its precious nucleotide into the extracellular milieu. However, growing evidence shows that this release is not a pure waste. It appears in fact that skeletal muscle cells utilize extracellular ATP to stimulate relevant autocrine/paracrine signaling pathways, related, for example, to the differentiative program. Present work was aimed at investigating the purinergic signaling of adult skeletal muscle. First, we confirmed, at the single fiber level, the release of ATP during contractile activity. The electrically stimulation (five 30-mV pulses of 2 ms duration with 500 ms interval) of cultured rat flexor digitorum brevis muscle fibers, isolated by the collagenase method, caused a substantial, rapid (seconds) release of ATP into the extracellular fluid, whose level returned to baseline with an estimated half-decay time of about 20 min. The decrease of the ATP concentration in the extracellular milieu represents an indirect observation of the presence of an ATP-hydrolyzing activity at muscle fiber surface, that we measured as high as 41 ± 4 nmol Pi liberated per mg protein (n = 3). Consistently, RT-PCR and WB analyses show the presence of NTPDase-1 and NTPDase-2, NPP1, and α-sarcoglycan in muscle fibers. Different purinergic receptors have been reported to be expressed in striated cells, a finding that we confirmed in adult muscle fibers. Moreover, we demonstrated that P2X4 is localized mainly in the T-tubule membranes, i.e., the critical site of excitation-contraction coupling of skeletal muscle.

Because all the elements of extracellular ATP signaling are present in skeletal muscle, we speculated that both the enhanced Ca2+ entry as well as the subsequent activation of Ca2+-dependent intracellular processes could modulate muscle contraction, especially during sustained contractile activity. Therefore, we examined different in vitro stimulation protocols on a typical slow-twitch muscle (soleus) in order to mimic such prolonged activity. We found that stimulation of soleus muscle at low frequencies (between 0.016 and 0.05 pulses/s) produced a slow progressive rise of twitch tension (potentiation) which, after 40 twitches, was about 20% higher than the initial value. We then applied different protocols devoted either to prevent or stimulate any possible effect of extracellular ATP in tension potentiation. The removal of extracellular ATP by specific enzymes (hexokinase/apyrase), the inhibition of P2X receptors by a cocktail of P2 blockers (PPADS, suramin, and RB-2), or Ca2+-free conditions, all abolished tension potentiation. The addition of Zn+ or ivermectin, at levels known to stimulate P2X4, were without effects, suggesting that the receptor is already fully activated. On the contrary, elevated doses of Zn+, ivermectin or 2meSATP reduced twitch tension potentiation.

Taking together, these data reveal that ATP-mediated Ca2+ entry plays, by still unknown mechanism/s, an important role in modulating the contractile activity of skeletal muscle and add new hints regarding the extracellular

ATP signaling of this tissue.

The work was funded by MIUR (PRIN03), Telethon Italy and CNR.

Extracellular interconversion of nucleotides indicates the presence of an ectoadenylate kinase activity in the rat hippocampus

Beáta Sperlágh and E. Sylvester Vizi

Department of Pharmacology, Institute of Experimental Medicine, Hungarian Academy of Sciences, Budapest, Hungary sperlagh@koki.hu Whereas the enzymatic mechanisms, responsible for the inactivation of extracellular ATP and giving rise to the formation of adenosine are well delineated in the nervous system, less is known about the reverse process i.e. the rephosporylation of nucleotides and/or nucleosides to ATP. Although an early study indicated that there is a catalytic activity in the extracellular space to convert 2 ADP to ATP and AMP in rat brain synaptosomes with Km in the millimolar range (Nagy et al., 1989) and similar ecto-adenylate kinase (EC activity has been reported in non-neuronal cells (Yegutkin et al., 2002), the reaction products of such an enzymatic conversion has not been directly demonstrated in intact brain preparations. To fulfill this objective, rat hippocampal slices and synaptosomes were exposed to increasing concentrations of various nucleotides and nucleosides and their extracellular interconversion were analysed by the HPLC-UV technique. If the slices were exposed to ATP (20–100–500 µM), it converted to ADP, AMP, adenosine, inosine and hypoxanthine according to the enzymatic process catalysed by the known families of ectonucleotidases. On the other hand, when the slices were challenged with ADP, in addition to the generation of adenosine, inosine and hypoxanthine, a parallel formation of ATP and AMP was also observed, which was strictly dependent on the initial concentration of ADP. A similar process was also observed in purified synaptosomes; however AMP formation in this case prevailed, and ATP generation could be only observed at higher initial ADP concentrations (100–500 µM). In the presence of the specific adenylate kinase inhibitor diadenosine pentaphosphate (Ap5A, 200 µM), the decomposition of ADP and the parallel formation of ATP in the slices was significantly slowed down, although some additional ATP generation was also detected, which was probably due to the breakdown of Ap5A to ATP, or to the release of ATP by P2 receptor activation (Ballerini et al., 1996), homo- or heteroexchange (Sperlágh et al., 2003) or by other mechanism. Nevertheless, no substantial change in the interconversion of ADP to ATP and AMP was found in the presence of the nucleoside transport inhibitor dypiridamole (3 µM), the P2 receptor antagonist PPADS (30 µM) and the connexin hemichannel inhibitor flufenamic acid (50 µM). When slices were exposed to AMP or adenosine, ectokinase activity was not detected, i.e. neither ATP nor ADP formation was observed.

In summary, here we demonstrate the parallel formation of ATP and AMP from ADP in the rat hippocampus, indicating that there is an ecto adenylate kinase activity in this brain area. This activity, which functions in the micromolar range, and is only partly associated to nerve terminals might represent the missing link of the nucleotide utilizing enzyme chain present in the brain tissue.

Extracellular NAD+ activates the P2Y11 purinergic receptor

1Moreschi, I., 1Bruzzone, S., 2Nicholas, R.A., 1Zocchi, E. and 1De Flora, A.

1Department of Experimental Medicine, Section of Biochemistry, and Center of Excellence for Biomedical Research, University of Genova, Viale Benedetto XV/1, 16132 Genova, Italy; 2Department of Pharmacology, University of North Carolina, Chapel Hill, NC, USA. email: iliana.moreschi@unige.it

Extracellular β-NAD+ (NAD+e) has been implicated in the regulation of intracellular calcium concentration ([Ca2+]i) in many cell types and by means of different mechanisms (1, 2, 3, 4, 5, 6). In human granulocytes, micromolar NAD+e activates cell-specific functional responses, i.e. superoxide and NO generation and chemotaxis, by triggering the following signalling cascade: i) activation of adenylate cyclase (AC) and accumulation of cyclic AMP (cAMP), ii) activation of protein kinase A, iii) stimulation of ADP-ribosyl cyclase activity with subsequent overproduction of cyclic ADP-ribose (cADPR), a universal Ca2+ mobilizer through activation of ryanodine receptors, and iv) influx of extracellular Ca2+ (7). Here we demonstrate that the addition of millimolar NAD+e to human granulocytes promotes the intracellular production of both Inositol-1,4,5-trisphosphate (IP3) and cAMP, leading to a two-step [Ca2+]i elevation: an initial IP3-mediated Ca2+ release from intracellular stores, followed by a second cADPR-mediated sustained influx of extracellular Ca2+. The use of suramin, a broad inhibitor of the P2Y purinoceptors sub-family, abrogated the NAD+-induced increases of [IP3]i, [cAMP]i, [cADPR]i and [Ca2+]i, thus indicating a role for a phospholipase C (PLC)- and AC-coupled P2Y receptor in initiating the signalling pathway triggered by NAD+e. The P2Y11 receptor is the only known member of the P2Y receptor sub-family coupled to both PLC and AC (8). Therefore, we investigated Ca2+ responses to NAD+e in an hP2Y11-transfected 1321N1 astrocytoma cells (9), compared to native P2Y11-negative cells. Micromolar NAD+e induced a two-step elevation of [Ca2+]i in transfected cells, but not in the control cells, due to the enhanced intracellular production of IP3, cAMP and cADPR. Thus, our results strongly suggest that β-NAD+e is an endogenous agonist of the P2Y11 purinoceptor.

Extracellular Nucleotides Enhance EGF-Induced C-Fos Expression in Two Breast Cancer Cell Lines, by Driving the EGF Receptor Into a Detergent-Insoluble Membrane Fraction.

A. Gartland1, R.A. Hipskind2, and J.A. Gallagher3.

1Academic Unit of Bone Biology, The University of Sheffield, Sheffield, S10 2RX, UK. 2IGMM, CNRS, 1919 Route de Mende. 34293 Montpellier. Cedex 5, France. 3Human Bone Cell Research Group, Department of Human Anatomy and Cell Biology, The University of Liverpool, Liverpool, L69 3GE U.K. a.gartland@sheffield.ac.uk

There is increasing evidence that extracellular nucleotides play a pivotal role in cancer growth, metastasis, and pain. We have previously shown that extracellular nucleotides synergise with growth factors to induce expression of the proto-oncogene c-fos in various cell types. In this study, we have examined the effect of extracellular nucleotides on growth factor-induced signalling in two breast cancer cell lines, HS578T and T47D. Using RT-PCR we found expression of various P2Y and P2X receptor subtypes in both cell lines. Co-stimulation of these cells with extracellular nucleotides and EGF lead to induction of endogenous c-fos above that observed with either treatment alone. This effect was recapitulated using HS578T and T47D c-fos reporter cells and was more pronounced with EN that predominantly act via P2Y receptors. EGF-induced phosphorylation of p44/p42 MAPK was unaffected in either cell type following co-stimulation, although phosphorylation of AKT and ERK5 was effected. EGFR phosphorylation appeared diminished upon co-stimulation, however the levels of total EGFR were also diminished in cold 1% Triton lysates in both cell types. When lysates were prepared in denaturing hot SDS buffer no reduction in EGFR levels was observed. This suggests that extracellular nucleotides may lead to inclusion of the EGFR into a Triton-insoluble membrane fraction, possibly lipid rafts, which may explain facilitated signalling to the c-fos promoter. Synergy between extracellular nucleotides and growth factors may represent an important mechanism promoting the growth and spread of tumours. This may be particularly relevant for breast cancer metastasis to bone as bone cells constitutively release ATP, and in response to mechanical strain. This could play a key role in creating an environment propitious for breast cancer cell growth after metastasis.

Extracellular nucleotides mediate LPS-induced IL-8 release from monocytes and neutrophil migration.

Filip Kukulski, Fethia Ben Yebdri, Julie Lefebvre, Jean Sévigny.

Centre de recherche en Rhumatologie et Immunologie, Université Laval, Ste-Foy, Québec, Canada. filip.kukulski@crchul.ulaval.ca

Increasing evidence suggests that extracellular nucleotides may mediate some of the proinflammatory effects of gram negative bacteria lipopolysaccharides (LPS). For instance, P2Y6 receptor activation is required for the LPS-induced secretion of interleukin-8 (IL-8) by THP-1 cells. Knowing that peripheral blood leukocytes (monocytes, lymphocytes and neutrophils) and HUVEC also express P2Y6 mRNA and that they respond to LPS by the secretion of IL-8, we have tested whether extracellular nucleotides mediate IL-8 release by these cells. As expected, all cells analyzed released IL-8 in response to LPS. However, this release was nucleotide-dependent only in monocytes, as it was significantly inhibited by a nucleotide scavenger (apyrase) and P2Y6 receptor antagonists (MRS 2578 and reactive blue 2). Importantly, the former inhibition by apyrase was not due to the formation of adenosine. In further support for a role of P2Y6 in IL-8 release by monocytes, these cells secreted important amounts of IL-8 when stimulated with UDP.

IL-8 is a major human chemokine, so we next tested the chemotactic activity of conditioned media from treated monocytes on neutrophil migration. Using modified Boyden chambers with an endothelial monolayer, we observed that the media of LPS-stimulated monocytes recruited significantly more neutrophils than the corresponding media to which LPS were added simultaneously with apyrase. Conditioned media of UDP-stimulated monocytes were as effective to induce neutrophil migration as those of the cells treated with LPS. In addition, IL-8 neutralizing antibodies added to the media of LPS- or UDP-stimulated monocytes substantially decreased their chemotactic activity, demonstrating that IL-8 was the main chemokine responsible for cellular migration in these assays.

In conclusion, extracellular nucleotides have the ability to regulate neutrophil recruitment to inflamed tissues by mediating IL-8 release from human monocytes/macrophages.

Feed forward cycle of hypotonic stress-induced ATP release, purinergic receptor activation and growth stimulation of prostate cancer cells

Rajender Nandigama#, Manju Padmasekar#, Maria Wartenberg§, and Heinrich Sauer# From the Department of Physiology, Justus-Liebig-University Gie§en#, Germany and the Department of Cell Biology, GKSS Research Center, Teltow, Germany§ E -mail: n_rajenderreddy@yahoo.com

ATP is released in many cell types upon mechanical strain, the physiological function of extracellular ATP is largely unknown, however. Here we report that ATP released upon hypotonic stress stimulated prostate cancer cell proliferation, activated purinergic receptors, increased intracellular [Ca2+]i and initiated downstream signalling cascades that involved mitogen activated protein kinases (MAPKs) ERK1/2 and p38 as well as PI3-kinase. MAPK activation, the calcium response as well as induction of cell proliferation upon hypotonic stress were inhibited by preincubation with the ATP scavenger apyrase, indicating that hypotonic stress-induced signalling pathways are elicited by released ATP. Hypotonic stress increased prostaglandin E2 (PGE2) synthesis. Consequently, ATP release was inhibited by antagonists of PI3-kinase (LY294002 and wortmannin), phospholipase A2 (MAFP), cyclooxygenase-2 (COX-2) (indomethacin, etodolac, NS398) and 5,8,11,14-eicosatetraynoic acid (ETYA), which are involved in arachidonic acid metabolism. Furthermore, ATP release was abolished in the presence of the adenylate cyclase (AC) inhibitor MDL-12,330A, indicating regulation of ATP-release by cAMP. The hypotonic stress-induced ATP release was significantly blunted when the ATP-mediated signal transduction cascade was inhibited on different levels, i.e. purinergic receptors were blocked by suramin and PPADS, the Ca2+ response was inhibited upon chelation of intracellular Ca2+ by BAPTA, and ERK1,2 as well as p38 were inhibited by UO126 and SB203580, respectively. In summary our data demonstrate that hypotonic stress initiates a feed forward cycle of ATP release and purinergic receptor signalling resulting in proliferation of prostate cancer cells.

Fluorescent N2,N3-ɛ-Adenine Nucleotide Probes for the Exploration of P2Y-receptors and NTPDases1

Bilha Fischera, Einat Sharona, Sebastien Levesqueb, Mercedes N. Munkondab, Jean Sévignyb, Denise Eckec, and Georg Reiserc

a. Department of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel

b. Centre de recherche en Rhumatologie et Immunologie, Université Laval, Sainte-Foy, Québec, Canada. c. Institute for Neurobiochemistry, Faculty of Medicine, Otto von Guericke University, Leipziger Str. 44 D-39120 Magdeburg, Germany e-mail: bfischer@mail.biu.ac.il

The fields of P2Y-receptors and NTPDases lack analytical tools for the exploration of these proteins. Therefore, we designed a novel probe: N2,N3-etheno-adenosine-5′-triphosphate, (N2,N3-ε-ATP) for improving the adenine fluorescence characteristics ( and λmax) while preserving its H-bonding pattern, required for molecular recognition. Here, we describe four novel syntheses of the target (-nucelotide, and related analogues. These methods are short (3–4 step syntheses), facile, and provide the product regiospecifically and in a reasonable yield. In addition, we report on the spectral properties of N2,N3-ɛ-ATP, including absorption and emission spectra, and dependence of the spectral features on pH and polarity of the medium. Specifically, maximum emission of N2,N3-ɛ-ATP in water is observed at 420 nm ( 0.03, excitation at 290 nm). The biochemical properties of N2,N3-ɛ-ATP were evaluated with respect to P2Y1-receptor and NTPDase1 and 2. N2,N3-ε-ATP was found equipotent to ATP at the P2Y1-receptor, and was hydrolysed by NTPDase1 and 2 at about the same rate as ATP. However, substitution at the exocyclic amine of N2,N3-ɛ-ATP significantly reduced affinity to P2Y1-R and NTPDase1. These observations emphasize the importance of free N1 and N6 positions for the recognition of ATP-based probes by target proteins. Furtherm