Inosine, not adenosine, initiates endothelial glycocalyx degradation in cardiac ischemia and hypoxia

Ischemia/reperfusion and hypoxia/reoxygenation of the heart both induce shedding of the coronary endothelial glycocalyx. The processes leading from an oxygen deficit to shedding are unknown. An involvement of resident perivascular cardiac mast cells has been proposed. We hypothesized that either adenosine or inosine or both, generated by nucleotide catabolism, attain the concentrations in the interstitial space sufficient to stimulate A3 receptors of mast cells during both myocardial ischemia/reperfusion and hypoxia/reoxygenation. Isolated hearts of guinea pigs were subjected to either normoxic perfusion (hemoglobin-free Krebs-Henseleit buffer equilibrated with 95% oxygen), 20 minutes hypoxic perfusion (buffer equilibrated with 21% oxygen) followed by 20 minutes reoxygenation, or 20 minutes stopped-flow ischemia followed by 20 minutes normoxic reperfusion (n = 7 each). Coronary venous effluent was collected separately from so-called transudate, a mixture of interstitial fluid and lymphatic fluid appearing on the epicardial surface. Adenosine and inosine were determined in both fluid compartments using high-performance liquid chromatography. Damage to the glycocalyx was evident after ischemia/reperfusion and hypoxia/reoxygenation. Adenosine concentrations rose to a level of 1 μM in coronary effluent during hypoxic perfusion, but remained one order of magnitude lower in the interstitial fluid. There was only a small rise in the level during postischemic perfusion. In contrast, inosine peaked at over 10 μM in interstitial fluid during hypoxia and also during reperfusion, while effluent levels remained relatively unchanged at lower levels. We conclude that only inosine attains levels in the interstitial fluid of hypoxic and postischemic hearts that are sufficient to explain the activation of mast cells via stimulation of A3-type receptors.     

Detection and characterization of a nucleoside transport system in human fibroblast lysosomes

Lysosomes contain enzymatic activities capable of degrading nucleic acids to their constituent nucleosides, but the manner by which these degradation products are released from the lysosome is unknown. To investigate this process, human fibroblast lysosomes, purified on Percoll density gradients, were incubated with [3H]adenosine at pH 7.0, and the amount of adenosine taken up by the lysosomes was measured. Adenosine uptake by fibroblast lysosomes attained a steady state by 12 min at 37 degrees C and was unaffected by the presence of 2 mM MgATP or changes in pH from 5.0 to 8.0. An Arrhenius plot was linear with an activation energy of 12.9 kcal/mol and a Q10 of 2.0. Lysosomal adenosine uptake is saturable, displaying a Km of 9 mM at pH 7.0 and 37 degrees C. Various nucleosides and the nucleobase, 6-dimethylaminopurine, strongly inhibit lysosomal adenosine uptake, whereas neither D-ribose or nucleotide monophosphates have any significant effect upon lysosomal adenosine uptake. On a molar basis, purines are recognized more strongly than pyrimidines. Changing the nature of the nucleoside sugar from ribose to arabinose or deoxyribose has little effect on reactivity with this transport system. The known plasma membrane nucleoside transport inhibitors, dipyridamole and nitrobenzylthioinosine, inhibit lysosomal nucleoside transport at relatively low concentrations (25 microM) relative to the Km of 9 mM for lysosomal adenosine uptake. The half-times of [3H]inosine and [3H]uridine efflux from fibroblast lysosomes ranged from 6 to 8 min at 37 degrees C. Trans effects were not observed to be associated with either inosine or uridine exodus. In contrast to adenosine uptake, adenine primarily enters fibroblast lysosomes by a route not saturable by high concentrations of various nucleosides. In conclusion, the saturability of lysosomal adenosine uptake and its specific, competitive inhibition by other nucleosides indicate the existence of a carrier-mediated transport system for nucleosides within fibroblast lysosomal membranes.     

Adenosine deaminase 1 and concentrative nucleoside transporters 2 and 3 regulate adenosine on the apical surface of human airway epithelia: implications for inflammatory lung diseases

Adenosine is a multifaceted signaling molecule mediating key aspects of innate and immune lung defenses. However, abnormally high airway adenosine levels exacerbate inflammatory lung diseases. This study identifies the mechanisms regulating adenosine elimination from the apical surface of human airway epithelia. Experiments conducted on polarized primary cultures of nasal and bronchial epithelial cells showed that extracellular adenosine is eliminated by surface metabolism and cellular uptake. The conversion of adenosine to inosine was completely inhibited by the adenosine deaminase 1 (ADA1) inhibitor erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA). The reaction exhibited Km and Vmax values of 24 microM and 0.14 nmol x min(-1) x cm(-2). ADA1 (not ADA2) mRNA was detected in human airway epithelia. The adenosine/mannitol permeability coefficient ratio (18/1) indicated a minor contribution of paracellular absorption. Adenosine uptake was Na+-dependent and was inhibited by the concentrative nucleoside transporter (CNT) blocker phloridzin but not by the equilibrative nucleoside transporter (ENT) blocker dipyridamole. Apparent Km and Vmax values were 17 microM and 7.2 nmol x min(-1) x cm(-2), and transport selectivity was adenosine = inosine = uridine > guanosine = cytidine > thymidine. CNT3 mRNA was detected throughout the airways, while CNT2 was restricted to nasal epithelia. Inhibition of adenosine elimination by EHNA or phloridzin raised apical adenosine levels by >3-fold and stimulated IL-13 and MCP-1 secretion by 6-fold. These responses were reproduced by the adenosine receptor agonist 5'-(N-ethylcarboxamido)adenosine (NECA) and blocked by the adenosine receptor antagonist, 8-(p-sulfophenyl) theophylline (8-SPT). This study shows that adenosine elimination on human airway epithelia is mediated by ADA1, CNT2, and CNT3, which constitute important regulators of adenosine-mediated inflammation.     

Transport mechanisms in isolated plasma membranes. Nucleoside processing by membrane vesicles from mouse fibroblast cells grown in defined medium.

Plasma membrane vesicles were isolated from a subline of L929 mouse fibroblasts grown on defined medium in the absence of serum. These vesicles were not significantly contaminated by mitochondria or endoplasmic reticulum. The isolation procedure, a modification of that originally developed by McKeel and Jarett (McKeel, D.W., and Jarett, L. (1970) J. Cell Biol. 44, 417-432) employs mechanical homogenization in isotonic medium followed by differential centrifugation. The resultant plasma membrane vesicles take up radioactivity when exposed to uniformly labeled nucleosides. Two subfractions of the plasma membrane were isolated, distinguished by their differing activity of 5'-nucleotidase and (Na+,K+)-stimulated ATPase, two well known plasma membrane enzyme markers. Uptake of nucleoside radioactivity was extensively studied in one subfraction; it was linear with time and membrane concentration over ranges used for the studies. Apparent Km values for uptake of radioactivity from adenosine, inosine, and uridine were 7.1 +/- 26 muM, respectively. Uptake of radioactivity from all three nucleosides exhibits a broad pH optimum from pH 7 to pH 9, but falls off rapidly at lower pH. N-Ethylmaleimide was an effective inhibitor of uptake of radioactivity from all three nucleosides; uptake of radioactivity from uridine is more sensitive than uptake of radioactivity from the purine nucleosides. Adenosine inhibited uptake of radioactivity from inosine more than from uridine. Inosine inhibited the uptake of radioactivity from adenosine, but uridine did not. Caffeine and 6-methylaminopurine riboside (6-N-methyladenosine differentially inhibit uptake of radioactivity from adenosine and inosine, and thus the vesicles apparently possess seperate transport systems for uptake of radioactivity from purine nucleosides and from uridine.     

Concentration of Nucleosides and Related Compounds in Cerebral and Cerebellar Cortical Areas and White Matter of the Human Brain

1. Nucleosides potentially participate in the neuronal functions of the brain. However, their distribution and changes in their concentrations in the human brain is not known. For better understanding of nucleoside functions, changes of nucleoside concentrations by age and a complete map of nucleoside levels in the human brain are actual requirements.2. We used post mortem human brain samples in the experiments and applied a recently modified HPLC method for the measurement of nucleosides. To estimate concentrations and patterns of nucleosides in alive human brain we used a recently developed reverse extrapolation method and multivariate statistical analyses.3. We analyzed four nucleosides and three nucleobases in human cerebellar, cerebral cortices and in white matter in young and old adults. Average concentrations of the 308 samples investigated (mean±SEM) were the following (pmol/mg wet tissue weight): adenosine 10.3±0.6, inosine 69.5±1.7, guanosine 13.5±0.4, uridine 52.4±1.2, uracil 8.4±0.3, hypoxanthine 108.6±2.0 and xanthine 54.8±1.3. We also demonstrated that concentrations of inosine and adenosine in the cerebral cortex and guanosine in the cerebral white matter are age-dependent.4. Using multivariate statistical analyses and degradation coefficients, we present an uneven regional distribution of nucleosides in the human brain. The methods presented here allow to creation of a nucleoside map of the human brain by measuring the concentration of nucleosides in microdissected tissue samples. Our data support a functional role for nucleosides in the brain.     

Coronary metabolic adaptation restricted by endothelin in the dog heart.

Endothelin elicits long-lasting vasoconstriction in the coronary bed. This remarkable spastic response raises the question whether or not the metabolic adaptive mechanisms of the coronaries are activated under endothelin effect. The role of the compensatory mediators adenosine and inosine was investigated before and after intracoronary (i.c.) administration of endothelin-1 (ET-1, 1.0 nmol) using 1-min reactive hyperemia (RH) tests on in situ dog hearts (n=15) with or without blocking the ATP-sensitive potassium (K+(ATP)) channels by glibenclamide (GLIB, 1.0 micromol min(-1), i.c.). The release of adenosine and inosine via the coronary sinus was measured by HPLC during the first minute of RH. Endothelin-1 reduced baseline coronary blood flow (CBF) and RH response (hyperemic excess flow (EF) control vs. ET-1: 81.7+/-13.6 vs. 43.4+/-10.9 ml, P<0.01), while it increased the net nucleoside release (adenosine, control vs. ET-1: 58.9+/-20.4 vs. 113.7+/-39.4 nmol, P<0.05; inosine: 242.1+/-81.8 vs. 786.9+/-190.8 nmol, P<0.05). GLIB treatment alone did not change baseline CBF but also reduced RH significantly and increased nucleoside release (EF control vs. GLIB: 72.1+/-11.7 vs. 31.9+/-5.5 ml, P<0.01; adenosine: 18.8+/-4.6 vs. 63.0+/-24.8 nmol, P<0.05; inosine: 113.0+/-37.2 vs. 328.2+/-127.5 nmol, P<0.05). Endothelin-1 on GLIB-treated coronaries further diminished RH and increased nucleoside release (EF: 21.5+/-8.0 ml, P<0.05 vs. GLIB; adenosine: 75.3+/-28.1 nmol, NS; inosine: 801.9+/-196.6 nmol, P<0.05 vs. GLIB). The data show that ET-1 reduces metabolic adaptive capacity of the coronaries, and this phenomenon is due to decreased vascular responsiveness and not to the blockade of ischemic mediator release from the myocardium. The coronary effect of ET-1 may partially be dependent on K+(ATP) channels.     

Determination of ribonucleosides,deoxyribonucleosides, and purine and pyrimidine bases in adult rabbit cerebrospinal fluid and plasma

Purine and pyrimidine base and nucleoside levels were measured in adult rabbit cisternal CSF and plasma by reversed-phase high-performance liquid chromatography. The concentrations of bases, nucleosides, and nucleoside phosphates were similar in plasma and CSF except for the adenosine phosphates and uracil which were higher in the plasma. In plasma and CSF, adenosine levels were low (0.12 microM) and guanosine, deoxyadenosine, deoxyguanosine, and deoxyinosine were not detectable (less than 0.1 microM); inosine and xanthine concentrations were 1-2 microM and hypoxanthine concentrations were approximately 5 microM; uridine (approximately 8 microM), cytidine (2-3 microM), and thymidine, deoxyuridine, and deoxycytidine (0.5-1.4 microM) were easily detectable. In both plasma and CSF, guanine, and thymine were undetectable (less than 0.1 microM), adenine and cytosine were less than 0.2 microM, but uracil was present (greater than 1 microM). Adenosine, inosine, and guanosine phosphates were also detectable at low concentrations in CSF and plasma. These results are consistent with the hypothesis that purine deoxyribonucleosides are synthesized in situ in the adult rabbit brain. In contrast, pyrimidine deoxyribonucleosides and ribonucleosides, and purine and pyrimidine bases are available in the CSF for use by the brain.     

Purine accumulation in human fat cell suspensions. Evidence that human adipocytes release inosine and hypoxanthine rather than adenosine

Human adipocytes are of limited viability (7 +/- 2% release of lactate dehydrogenase/h) and contain active ectophosphatases which are capable of sequentially degrading ATP to adenosine. At densities of 30,000-40,000 cells/ml, human fat cell suspensions accumulated adenosine, inosine, and hypoxanthine, and their concentrations were 38 +/- 8, 120 +/- 10, and 31 +/- 7 nmol/liter after 3 h of incubation. Dipyridamole (10 mumol/liter), an inhibitor of nucleoside transport, caused a 5-7-fold increase in adenosine accumulation which was reduced by 85% on inhibition of ectophosphatases by beta-glycerophosphate and antibodies against ecto-5'-nucleotidase or alpha, beta-methylene 5'-adenosine diphosphate (10 mumol/liter), respectively, indicating that most of the adenosine is produced in the extracellular compartment. Accordingly, the spontaneous accumulation of adenosine was reduced beyond 5 nmol/liter on inhibition of ectophosphatase activities or removal of extracellular AMP by AMP deaminase (4 units/ml). Added adenosine (30 nmol/liter) disappeared until its concentration approached 5 nmol/liter. Isoproterenol (1 mumol/liter) had no effect on adenosine accumulation regardless whether purine production from extracellular sources was minimized or not. In contrast to adenosine, the concentrations of inosine and hypoxanthine displayed only a modest decrease (30-50%) on inhibition of ectophosphatase activities. In addition, isoproterenol caused a 2-3-fold increase in inosine and hypoxanthine production which was concentration-dependent and could be inhibited by propranolol. It is concluded that the adenosine that accumulates in human adipocyte suspensions is almost exclusively derived from adenine nucleotides which are released by leaking cells. By contrast, inosine and hypoxanthine are produced inside the cells, and the release of these latter purines appears to be linked to ATP turnover via adenylate cyclase.     

Enzymatic Synthesis of 5-Methyluridine from Adenosine and Thymine with High Efficiency

5-Methyluridine (5MU) was synthesized efficiently from adenosine, thymine, and phosphate by a combination of adenosine deaminase (ADA), purine nucleoside phosphorylase (PUNP), pyrimidine nucleoside phosphorylase (PYNP), and xanthine oxidase (XOD). Adenosine was converted into inosine first by ADA. 5MU and hypoxanthine were synthesized from inosine and thymine by PUNP and PYNP. The hypoxanthine formed was converted into urate via xanthine by XOD. After inosine was completely consumed, an equilibrium state, in which 5MU, thymine, ribose-1-phosphate, and phosphate were involved, was achieved. At the equilibrium state, the maximum yield of 5MU was obtained. The yield of 5MU was 74%, when the initial concentrations of adenosine, thymine, and phosphate were 5 mM each. On the other hand, in the absence of ADA or XOD the yield of 5MU was 1.8%. Several kinds of nucleosides were also synthesized with high yield by the same method.     

Passive and carrier-mediated permeation of different nucleosides through the reconstituted nucleoside transporter

When reconstituted into proteoliposomes, the human erythrocyte nucleoside transporter catalysed nitrobenzylthioguanosine (NBTGR)-sensitive zero-trans influx of three different nucleosides at broadly similar rates (inosine, uridine greater than adenosine). However, proteoliposomes also exhibited high rates of NBTGR-insensitive uptake of adenosine, making this nucleoside unsuitable for reconstitution studies. Equivalent high rates of adenosine influx were observed in protein-free liposomes, establishing that this permeability pathway represents simple diffusion of nucleoside across the lipid bilayer. In contrast to adenosine, inosine and uridine exhibited acceptable rates of NBTGR-insensitive uptake. Of the two, inosine is the more attractive permeant for reconstitution experiments, having a 2.5-fold lower basal membrane permeability. Studies of nucleoside transport specificity in reconstituted membrane vesicles should take account of the widely different passive permeabilities of different nucleosides.     

Elucidation of aberrant purine metabolism: application to hypoxanthine-guanine phosphoribosylstransferase- and adenosine kinase-deficient mutants, and IMP dehydrogenase- and adenosine deaminase-inhibited human lymphoblasts

We propose that the ratio of [14C]formate-labelled purine nucleosides and bases (both intra and extracellular) to nucleic acid purines provides, in exponentially growing cultures, a sensitive index for comparative studies of purine metabolism. This ratio was 4-fold greater for an HGPRT- mutant than for the parental HGPRT+ human lymphoblast line. The major components of the labelled nucleoside and base fraction were hypoxanthine and inosine. By blocking adenosine deaminase activity with coformycin we found that approx. 90% of inosine was formed directly from IMP rather than the route IMP leads to AMP leads to adenosine leads to inosine. The ratio of labelled base + nucleosides to nucleic acids was essentially unchagned for an AK- lymphoblast line and 2-fold greater than control for an HGPRT(-)-KAK- line, demonstrating that a deficiency of adenosine kinase alone has little effect on the accumulation of purine nucleosides and bases. Although adenosine was a minor component of the nucleoside and base fraction, the adenosine fraction increased from 3 to 13% with the addition of coformycin to the HGPRT(-)-AK- line. In the parental and HGPRT- lines, adenosine was shown to be primarily phosphorylated rather than deaminated at concentrations less than 5 microM. Inhibition of IMP dehydrogenase activity by mycophenolic acid caused a 12- and 3-fold increase in the rate of production of labelled base and nucleoside in the parent and HGPRT- cells respectively. These results suggest that a mutationally induced partial deficiency in the activities converting IMP to guanine nucleotides may result in an increased catabolism of IMP.     

Features of adenosine metabolism of mouse heart

Adenosine metabolism and transport were evaluated in the isolated perfused mouse heart and compared with the well-established model of isolated perfused guinea pig heart. Coronary venous release of adenosine under well-oxygenated conditions in the mouse exceeds that in the guinea pig threefold when related to tissue mass. Total myocardial adenosine production rate under this condition was approximately 2 nmol/min per gramme and similar in both species. Coronary resistance vessels of mice are highly sensitive to exogenous adenosine, and the threshold for adenosine-induced vasodilation is approximately 30 nmol/l. Adenosine membrane transport was largely insensitive to nitrobenzyl-thioinosine (NBTI) in mouse heart, which is in contrast to guinea pig and several other species. This indicates the dominance of NBTI-insensitive transporters in mouse heart. For future studies, the assessment of cytosolic and extracellular adenosine metabolism and its relationship with coronary flow will require the use of more effective membrane transport blockers.     

Genetic analysis of nucleoside transport in Leishmania donovani.

Genetic dissection of nucleoside transport in Leishmania donovani indicates that the insect vector form of these parasites possesses two biochemically distinct nucleoside transport systems. The first transports inosine, guanosine, and formycin B, and the second transports pyrimidine nucleosides and the adenosine analogs, formycin A and tubercidin. Adenosine is transported by both systems. A mutant, FBD5, isolated by virtue of its resistance to growth inhibition by 5 microM formycin B, cannot efficiently transport inosine, guanosine, or formycin B. This cell line is also cross-resistant to growth inhibition by a spectrum of cytotoxic analogs of inosine and guanosine. A second parasite mutant, TUBA5, isolated for its resistance to 20 microM tubercidin, cannot take up from the culture medium radiolabeled tubercidin, formycin A, uridine, cytidine, or thymidine. Both the FBD5 and the TUBA5 cell lines have about a 50% reduced capacity to take up adenosine, indicating that adenosine is transported by both systems. A tubercidin-resistant clonal derivative of FBD5, FBD5-TUB, has acquired the combined biochemical phenotype of each single mutant. The wild-type and mutant cell lines transport purine bases and uracil with equal efficiency. Mutational analysis of the relative growth sensitivities to cytotoxic nucleoside analogs and the selective capacities to take up exogenous radiolabeled nucleosides from the culture medium have enabled us to define genetically the multiplicity and substrate specificities of the nucleoside transport systems in L. donovani promastigotes.     

Kinetic characterization of adenosine deaminase activity in zebrafish (Danio rerio) brain

Adenosine deaminase (ADA; EC 3.5.4.4) activity is responsible for cleaving adenosine to inosine. In this study we described the biochemical properties of adenosine deamination in soluble and membrane fractions of zebrafish (Danio rerio) brain. The optimum pH for ADA activity was in the range of 6.0-7.0 in soluble fraction and reached 5.0 in brain membranes. A decrease of 31.3% on adenosine deamination in membranes was observed in the presence of 5 mM Zn(2+), which was prevented by 5 mM EDTA. The apparent K(m) values for adenosine deamination were 0.22+/-0.03 and 0.19+/-0.04 mM for soluble and membrane fractions, respectively. The apparent V(max) value for soluble ADA activity was 12.3+/-0.73 nmol NH(3) min(-1) mg(-1) of protein whereas V(max) value in brain membranes was 17.5+/-0.51 nmol NH(3) min(-1) mg(-1) of protein. Adenosine and 2'-deoxyadenosine were deaminated in higher rates when compared to guanine nucleosides in both fractions. Furthermore, a significant inhibition on adenosine deamination in both soluble and membrane fractions was observed in the presence of 0.1 mM of erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA). The presence of ADA activity in zebrafish brain may be important to regulate the adenosine/inosine levels in the CNS of this species.     

The emerging role of adenosine deaminases in insects

Adenosine deaminases catalyze the deamination of adenosine and deoxyadenosine into their respective inosine nucleosides. Recent sequencing of the genomes of several model organisms and human reveal that Metazoa usually have more than one adenosine deaminase gene. A deficiency in the gene encoding the major enzyme is lethal in mouse and Drosophila and leads to severe combined deficiency (SCID) in human. In these organisms, enzyme deficiency causes increased adenosine/deoxyadenosine concentration in body fluids and some organs. Elevated levels of adenosine and deoxyadenosine are toxic to certain mammalian and insect cells, and it was shown for human and mouse that it is a primary cause of pathophysiological effects. Data suggest that the major role of adenosine deaminases in various taxa is the protection of tissues against increased levels of adenosine and deoxyadenosine. This review also discusses potential roles of adenosine deaminases in Drosophila metamorphosis and the employment of a Drosophila model to study the cell-specific toxicity of elevated nucleoside levels.     

Use of the integrated steady state rate equation to investigate product inhibition of human red cell adenosine deaminase and its relevance to immune dysfunction

The analysis of progress curves using the integrated rate equation was applied to the adenosine deaminase-catalyzed conversion of adenosine to inosine. Adenosine deaminase was purified from human red blood cells of phenotypes ADA 1, ADA 2, and ADA 2-1. For all three types, no measurable product inhibition by inosine was observed. These results do not confirm the hypothesis that inosine accumulation in purine nucleoside phosphorylase deficiency causes adenosine deaminase inhibition, resulting in a common mechanism for the immune defects related to these two enzyme deficiencies.     

Adenosine transport by the lung

Adenosine, a nucleoside and potent vasodilator, has been found to be taken up by the lung and converted by deamination into inosine and hypoxanthine. In a single circulation through an isolated rat lung, 69.3 +/- 3.3% of infused [14C]adenosine (10 microM) was removed from the circulation. Uptake of [14C]adenosine remained unchanged when deamination of adenosine was inhibited by 8-azaguanine or coformycin. In a single passage of adenosine through the pulmonary artery, very little of the deaminated products appeared in the pulmonary circulation, but when adenosine was recirculated through the pulmonary circulation inosine and hypoxanthine appeared in the venous effluent. These adenosine metabolites were also taken up by the lung. A major portion of the circulating adenosine was transported into the lung, where it was used to synthesize adenine nucleotides. Inhibition of adenosine kinase by iodotubercidin resulted in reduced formation of ATP and ADP. Uptake of adenosine by the lung was saturable on a concentration gradient and was a passive process because it was not affected by the absence of glucose or the presence of ouabain. Km and Vmax for adenosine transport were 0.227 mM and 4.6 mumol.min-1.g lung-1, respectively. Adenosine transport was inhibited by adenosine analogues, and the inhibitions were found to be competitive in nature. These results suggest that a specific and rate-limiting transport system exists in the lung for adenosine.     

Transport of nucleosides across the Plasmodium falciparum parasite plasma membrane has characteristics of PfENT1

Like all parasitic protozoa, the human malaria parasite Plasmodium falciparum lacks the enzymes required for de novo synthesis of purines and it is therefore reliant upon the salvage of these compounds from the external environment. P. falciparum equilibrative nucleoside transporter 1 (PfENT1) is a nucleoside transporter that has been localized to the plasma membrane of the intraerythrocytic form of the parasite. In this study we have characterized the transport of purine and pyrimidine nucleosides across the plasma membrane of 'isolated' trophozoite-stage P. falciparum parasites and compared the transport characteristics of the parasite with those of PfENT1 expressed in Xenopus oocytes. The transport of nucleosides into the parasite: (i) was, in the case of adenosine, inosine and thymidine, very fast, equilibrating within a few seconds; (ii) was of low affinity [K(m) (adenosine) = 1.45 +/- 0.25 mM; K(m) (thymidine) = 1.11 +/- 0.09 mM]; and (iii) showed 'cross-competition' for adenosine, inosine and thymidine, but not cytidine. The kinetic characteristics of nucleoside transport in intact parasites matched very closely those of PfENT1 expressed in Xenopus oocytes [K(m) (adenosine) = 1.86 +/- 0.28 mM; K(m) (thymidine) = 1.33 +/- 0.17 mM]. Furthermore, PfENT1 transported adenosine, inosine and thymidine, with a cross-competition profile the same as that seen for isolated parasites. The data are consistent with PfENT1 serving as a major route for the uptake of nucleosides across the parasite plasma membrane.     

Stimulus- and cell-type-specific release of purines in cultured rat forebrain astrocytes and neurons

Adenosine is formed during conditions that deplete ATP, such as ischemia. Adenosine deaminase converts adenosine into inosine, and both adenosine and inosine can be beneficial for postischemic recovery. This study investigated adenosine and inosine release from astrocytes and neurons during chemical hypoxia or oxygen-glucose deprivation. In both cell types, 2-deoxyglucose was the most effective stimulus for depleting cellular ATP and for evoking inosine release; in contrast, oxygen-glucose deprivation evoked the greatest adenosine release. alpha,beta-Methylene ADP, an inhibitor of ecto-5'nucleotidase, significantly reduced adenosine release from astrocytes but not neurons. Dipyridamole, an inhibitor of equilibrative nucleoside transporters, inhibited both adenosine and inosine release from neurons. Erythro-9-(2-hydroxy-3-nonyl)adenine, an inhibitor of adenosine deaminase, reduced neuronal inosine release evoked by oxygen-glucose deprivation but not by 2-deoxyglucose treatment. These data indicate that (1). astrocytes release adenine nucleotides that are hydrolyzed extracellularly to adenosine, whereas neurons release adenosine per se, (2). inosine is formed intracellularly and released via nucleoside transporters, and (3). inosine is formed by an adenosine deaminase-dependent pathway during oxygen-glucose deprivation but not during 2-deoxyglucose treatment. In summary, the metabolic pathways for adenosine formation and release were cell-type dependent whereas the pathways for inosine formation were stimulus dependent.     

Role of S-adenosylhomocysteine hydrolase in adenosine metabolism in mammalian heart.

S-Adenosylhomocysteine hydrolase of mammalian hearts from different species is exclusively a cytosolic enzyme. The apparent Km for the guinea-pig enzyme was 2.9 microM (synthesis) and 0.39 microM (hydrolysis). Perfusion of isolated guinea-pig hearts for 120 min with L-homocysteine thiolactone (0.23 mM) and adenosine (0.1 mM), in the presence of erythro-9-(2-hydroxynon-3-yl)adenine to inhibit adenosine deaminase, caused tissue contents of S-adenosylhomocysteine to increase from 3.5 to 3600 nmol/g. When endogenous adenosine production was accelerated by perfusion of hearts with hypoxic medium (30% O2), L-homocysteine thiolactone (0.23 mM) increased S-adenosyl-homocysteine 17-fold to 64.3 nmol/g within 15 min. In the presence of 4-nitro-benzylthioinosine (5 microM), an inhibitor of adenosine transport, S-adenosylhomocysteine further increased to 150 nmol/g. L-Homocysteine thiolactone decreased the hypoxia-induced augmentation of adenosine, inosine and hypoxanthine in the tissue and the release of these purines into the coronary system by more than 50%. Our findings indicate that L-homocysteine can profoundly alter adenosine metabolism in the intact heart by conversion of adenosine into S-adenosylhomocysteine. Adenosine formed during hypoxia was most probably generated within the myocardial cell.