Asparagine 81, an invariant glycosylation site near apyrase conserved region 1, is essential for full enzymatic activity of ecto-nucleoside triphosphate diphosphohydrolase 3

N-linked glycosylation is important for the function, cellular localization, and oligomerization of membrane-bound ecto-nucleoside triphosphate diphosphohydrolases (eNTPDases). NTPDase3 is a prototypical cell membrane-associated eNTPDase, which is equally related and enzymatically intermediate to the other two cell surface membrane NTPDases (NTPDase1 and 2). The protein sequence of NTPDase3 contains seven putative N-glycosylation sites located in the ecto-domain. Only one of these putative glycosylation sites, asparagine 81 in NTPDase3, which is located near apyrase conserved region 1 (ACR1), is invariant in all the cell surface membrane eNTPDases. Using site-directed mutagenesis, mutants were constructed to eliminate this highly conserved N-glycosylation site in NTPDase3. The results indicate that glycosylation at this position is essential for full enzymatic activity, with mutant ATPase activity decreased more than ADPase activity. Enzymatic deglycosylation of this site is shown to be responsible for the inactivation of the wild-type enzyme by treatment with peptide N-glycosidase-F. In addition, glycosylation of this conserved site is necessary for the stabilization/stimulation of nucleotidase activity upon treatment with the lectin concanavalin A. However, lack of glycosylation at this site did not result in large changes in tertiary or quaternary structure, as measured by Cibacron blue binding, chemical cross-linking, and native gel electrophoretic analysis. Since this N-glycosylation site is invariant in cell membrane eNTPDases, it is postulated that glycosylation of this residue near ACR1 is crucial for full enzymatic activity of the cell membrane NTPDases.     

NTPDase and 5′-nucleotidase activities in platelets of human pregnants with a normal or high risk for thrombosis

The nucleotide degrading enzymes, ectonucleotidases, present on the platelet surface of human pregnant with a normal (without complications) or high risk for thrombosis (hypertension and gestational diabetes) were studied. NTPDase (E.C. 3.6.1.5, CD39) and 5′-nucleotidase (E.C. 3.1.3.5, CD73) activities of four patient groups, non-pregnant (NP, n = 18), pregnant without complications (P, n = 25), pregnant with hypertension (HP, n = 15) and pregnant with gestational diabetes mellitus (GDP, n = 10), were analyzed. Increased NTPDase activities were observed in the groups P (37.0%, S.D. = 2.03 and 34.0%, S.D. = 3.19), HP (40.0%, S.D. = 3.32 and 56.0%, S.D. = 3.25) and GDP (23.0%, S.D. = 2.30 and 42.0%, S.D. = 2.26) in comparison to the control group NP (p < 0.01, S.D. = 1.92 and S.D. = 2.48) when ATP and ADP were used as substrate, respectively. AMP was used as substrate to determine the 5′-nucleotidase activities, which showed to be elevated in the groups P (45.0%, S.D. = 1.73), HP (54.0%, S.D. = 2.64) and GDP (68.0%, S.D. = 1.69) when compared to the control group NP (p < 0.01, S.D. = 1.26). However, no statistically significant differences were observed between the groups P, HP and GDP. As a consequence, the enhanced ATP, ADP and AMP hydrolysis was ascribed to the pregnancy itself, independent of a normal or high risk for thrombosis. The enhanced NTPDase and 5′-nucleotidase activities in platelets suggest that these enzymes are involved in the thromboregulation process in the pregnancy.     

Soluble NTPDase: An additional system of nucleotide hydrolysis in rat blood serum

The participation of a nucleoside triphosphate diphosphohydrolase in the nucleotide hydrolysis by rat blood serum was evaluated. Nucleoside triphosphate diphosphohydrolase and phosphodiesterase are enzymes possibly involved in ATP and ADP hydrolysis. The specific activity of the phosphodiesterase activity (using thymidine 5'-monophosphate p-nitrophenyl ester as substrate) was 4.92 +/- 0.73 (mean +/- SD, n = 10) nmol p-nitrophenol.min(-1).mg(-1) protein and the specific activities for ATP and ADP were 1.31 +/- 0.37 (mean +/- SD, n = 7) and 1.36 +/- 0.25 (mean +/- SD, n = 5) nmol Pi.min(-1).mg(-1) protein, respectively. A competition plot demonstrated that ATP and ADP hydrolysis occurs at the same active site. The effect of suramin and phenylalanine on ATP, ADP and thymidine 5'-monophosphate p-nitrophenyl ester hydrolysis was investigated. The results were opposite considering the hydrolysis of ATP and ADP and that of the substrate marker for the enzyme phosphodiesterase. These results are indicative of the presence of, at least, two enzymes participating in the serum nucleotide hydrolysis. The presence of cAMP did not affect the hydrolysis velocity of ATP and ADP, while thymidine 5'-monophosphate p-nitrophenyl ester hydrolysis was inhibited by cAMP by approximately 47%, suggesting that the hydrolysis of the ATP and ADP, under our assay conditions, occurs at a different site from the phosphodiesterase site. Both enzyme activities, in the rat blood serum, may be involved in the modulation of the nucleotide/nucleoside ratio in the circulation, serving an in vivo homeostatic and antithrombotic function. In addition, the phosphodiesterase may act on DNA or RNA liberated upon tissue injury and/or cell death.     

Preferential Release of ATP and Its Extracellular Catabolism as a Source of Adenosine upon High- but Not Low-Frequency Stimulation of Rat Hippocampal Slices

Abstract: The release of adenosine and ATP evoked by electrical field stimulation in rat hippocampal slices was investigated with the following two patterns of stimulation: (1) a brief, high-frequency burst stimulation (trains of stimuli at 100 Hz for 50 ms applied every 2 s for 1 min), to mimic a long-term potentiation (LTP) stimulation paradigm, and (2) a more prolonged (3 min) and low-frequency (5 Hz) train stimulation, to mimic a long-term depression (LTD) stimulation paradigm. The release of ATP was greater at a brief, high-frequency burst stimulation, whereas the release of [³H]adenosine was slightly greater at a more prolonged and low-frequency stimulation. To investigate the source of extracellular adenosine, the following two pharmacological tools were used; α,β-methylene ADP (AOPCP), an inhibitor of ecto-5′-nucleotidase, to assess the contribution of the catabolism of released adenine nucleotides as a source of extracellular adenosine, and S-(4-nitrobenzyl)-6-thioinosine (NBTI), an inhibitor of adenosine transporters, to assess the contribution of the release of adenosine, as such, as a source of extracellular adenosine. At low-frequency stimulation, NBTI inhibited by nearly 50% the evoked outflow of [³H]adenosine, whereas AOPCP inhibited [³H]adenosine outflow only marginally. In contrast, at high-frequency stimulation, AOPCP inhibited by 30% the evoked release of [³H]adenosine, whereas NBTI produced a 40% inhibition of [³H]adenosine outflow. At both frequencies, the kinetics of evoked [³H]adenosine outflow was affected in different manners by AOPCP and NBTI; NBTI mainly depressed the rate of evoked [³H]adenosine outflow, whereas AOPCP mainly inhibited the later phase of evoked [³H]adenosine accumulation. These results show that there is a simultaneous, but quantitatively different, release of ATP and adenosine from rat hippocampal slices stimulated at frequencies that can induce plasticity phenomena such as LTP (100 Hz) or LTD (5 Hz). The source of extracellular adenosine is also different according to the frequency of stimulation; i.e., at a brief, high-frequency stimulation there is a greater contribution of released adenine nucleotides for the formation of extracellular adenosine than at a low frequency with a more prolonged stimulation.     

Phospholipase C axis is the preferential pathway leading to PKC activation following PTH or PTHrP stimulation in human term placenta

In the present report we describe an NTPDase 1 (ATP diphosphohydrolase; ecto-apyrase; EC 3.6.1.5) in rat hippocampal slices. The effect of glutamate on the ATPase and ADPase activities in rat hippocampal slices of different ages was also studied since adenosine, the final product of an enzymatic chain that includes NTPDase 1 and 5'-nucleotidase, can act upon A1 receptors in turn decreasing the release of glutamate. Hippocampal slices from 7, 14, 20-23 and 60 day-old rats were prepared and ATPase and ADPase activities were measured. The parallelism of ATPase and ADPase activities in all parameters tested indicated the presence of an ATP diphosphohydrolase. In addition, a Chevillard plot indicated that ATP and ADP are hydrolyzed at the same active site on the enzyme. ATPase activity was significantly activated by glutamate in 20-23 and 60 day-old rats, but ADPase activity was not activated. These results could indicate distinct behavior of the ATPase and ADPase activities of NTPDase 1 in relation to glutamate or the simultaneous action of the ecto-ATPase. Activation of ATPase activity by glutamate may constitute an important role in this developmental period, possibly protecting against the neurotoxicity induced by ATP, as well as producing high levels of ADP, by increasing adenosine production, a neuroprotective compound.     

Transmembrane domain interactions affect the stability of the extracellular domain of the human NTPDase 2

Human NTPDase2 and chicken NTPDase8 are cell surface nucleotidases that contain two transmembrane domains (TMD) and five apyrase conserved regions (ACRs). ACR1 is located near the N-terminal TMD whereas ACR5 is located near the C-terminal TMD. The human NTPDase2 activity is decreased by low concentration of NP-40 and at temperatures higher than 37 °C, and undergoes substrate inactivation, whereas the chicken NTPDase8 activity is not. When freed from membrane anchorage, the soluble human NTPDase2 is no longer inactivated by detergents, high temperature, and substrate. These characteristics are retained in the hu-ck ACR1,5 chimera in which the extracellular domain is anchored to the membrane by the two TMDs of the chicken NTPDase8. The hu-ck ACR1,5 chimera is the first chimeric NTPDase reported that shows a resistance to membrane perturbation and substrate inactivation. Our results indicate that the strengths of interaction of the respective TMD pairs of the human NTPDase2 and chicken NTPDase8 determine their different responses to membrane perturbation and substrate.     

Extracellular Interconversion of Nucleotides Reveals an Ecto-Adenylate Kinase Activity in the Rat Hippocampus

Here, the extracellular interconversion of nucleotides and nucleosides was investigated in rat hippocampal slices and synaptosomes by an HPLC-UV technique. Adenosine 5′-triphosphate (ATP) was converted to adenosine 5′-diphosphate (ADP), adenosine 5′-monophosphate (AMP), adenosine, inosine, and hypoxanthine in the slices, whereas ADP elicited parallel and concentration-dependent formation of ATP and AMP. The specific adenylate kinase inhibitor diadenosine pentaphosphate decreased the rate of decomposition of ADP and inhibited the formation of ATP. No substantial changes in the interconversion of ADP to ATP and AMP were found in the presence of dipyridamole, flufenamic acid, the P2 receptor antagonist pyridoxal-5-phosphate-6-azophenyl-2′,4′-disulphonic acid tetrasodium (PPADS), and the alkaline phosphatase substrate para-nitrophenylphosphate. Negligible levels of nucleotides were generated when uridine 5′-diphosphate (UDP), AMP or adenosine were used as substrates. Ecto-adenylate kinase activity was also observed in purified synaptosomes. In summary, we demonstrate the presence of an ecto-adenylate kinase activity in the hippocampus, which is a previously unrecognized pathway that influences the availability of purines in the central nervous system.     

Negative feedback of extracellular ADP on ATP release in goldfish hepatocytes: A theoretical study

A mathematical model was built to account for the kinetic of extracellular ATP (ATPe) and extracellular ADP (ADPe) concentrations from goldfish hepatocytes exposed to hypotonicity. The model was based on previous experimental results on the time course of ATPe accumulation, ectoATPase activity, and cell viability [Pafundo et al., 2008].The kinetic of ATPe is controlled by a lytic ATP flux, a non-lytic ATP flux, and ecto-ATPase activity, whereas ADPe kinetic is governed by a lytic ADP flux and both ecto-ATPase and ecto-ADPase activities. Non-lytic ATPe efflux was included as a diffusion equation modulated by ATPe activation (positive feedback) and ADPe inhibition (negative feedback).The model yielded physically meaningful and stable steady-state solutions, was able to fit the experimental time evolution of ATPe and simulated the concomitant kinetic of ADPe. According to the model during the first minute of hypotonicity the concentration of ATPe is mainly governed by both lytic and non-lytic ATP efflux, with almost no contribution from ecto-ATPase activity. Later on, ecto-ATPase activity becomes important in defining the time dependent decay of ATPe levels. ADPe inhibition of the non-lytic ATP efflux was strong, whereas ATPe activation was minimal. Finally, the model was able to predict the consequences of partial inhibition of ecto-ATPase activity on the ATPe kinetic, thus emulating the exposure of goldfish cells to hypotonic medium in the presence of the ATP analog AMP-PCP. The model predicts this analog to both inhibit ectoATPase activity and increase non-lytic ATP release.     

Kinetics of Extracellular ATP from Goldfish Hepatocytes: A Lesson from Mathematical Modeling

In goldfish hepatocytes, hypotonic exposure leads to cell swelling, followed by a compensatory shrinkage termed RVD. It has been previously shown that ATP is accumulated in the extracellular medium of swollen cells in a non-linear fashion, and that extracellular ATP (ATPe) is an essential intermediate to trigger RVD. Thus, to understand how RVD proceeds in goldfish hepatocytes, we developed two mathematical models accounting for the experimental ATPe kinetics reported recently by Pafundo et al. in Am. J. Physiol. 294, R220–R233, 2008. Four different equations for ATPe fluxes were built to account for the release of ATP by lytic (J _L ) and nonlytic mechanisms (J _NL ), ATPe diffusion (J _D ), and ATPe consumption by ectonucleotidases (J _V ). Particular focus was given to J _NL , defined as the product of a time function (J _R ) and a positive feedback mechanism whereby ATPe amplifies J _NL . Several J _R functions (Constant, Step, Impulse, Gaussian, and Lognormal) were studied. Models were tested without (model 1) or with (model 2) diffusion of ATPe. Mathematical analysis allowed us to get a general expression for each of the models. Subsequently, by using model dependent fit (simulations) as well as model analysis at infinite time, we observed that:

Effects of phenylalanine and phenylpyruvate on ATP-ADP hydrolysis by rat blood serum

The nucleotide (ATP-ADP)/nucleoside (adenosine) ratio in the circulation can modulate the processes of vasoconstriction, vasodilatation and platelet aggregation. The main objective of the present study with rat blood serum was to evaluate the possibility of changes in nucleotide hydrolysis by phenylalanine (Phe) and phenylpyruvate (PP), the levels of which could increase in the circulation of individuals with phenylketonuria. Results demonstrated that Phe in the range 1.0-5.0 mM inhibited the ADP hydrolysis by rat serum. The effect of inhibition by Phe on ATP hydrolysis appeared only at a concentration of 5.0 mM. PP had no significant effect upon nucleotide hydrolysis. Kinetic analysis indicated that the inhibition of ADP and ATP hydrolysis by Phe in rat blood serum is uncompetitive. Conversely, Phe and PP did not affect the hydrolysis of p-nitrophenyl-5'-TMP by rat serum.     

A case of serendipity*

An account is given of how a sensitive bioassay system for measurement of the neurotransmitter acetylcholine serendipitously led to the identification of adenosine triphosphate (ATP) released in vitro from active skeletal muscle. Subsequent application of the identification procedures to exercising human muscle in vivo, cardiac muscle cells in vitro, and human erythrocytes exposed to hypoxia gave rise to the general concept of ATP as a molecule that could influence cell function from the extracellular direction. Mechanisms of ATP release from cells in terms of "trigger" events such as mechanical distortion of the membrane, depolarization of the membrane, and exposure to hypoxia are discussed. Potential therapeutic uses of extracellular ATP in cancer therapy, radiation therapy, and a possible influence upon aging are discussed. Possible roles (distant and local) of extracellular ATP released from muscle during whole body exercise are discussed.     

Effects of extracellular ATP on local and systemic responses of bean (Phaseolus vulgaris L) leaves to wounding

Wounding increased the extracellular Adenosine 5?-triphosphate (eATP) level of kidney bean leaves. Treatment with wounding or exogenous ATP increased the hydrogen peroxide (H₂O₂) content, activities of catalase and polyphenol oxidase, and malondialdehyde content in both the treated and systemic leaves. Pre-treatment with ATP-degrading enzyme, apyrase, to the wounded leaves reduced the wound-induced local and systemic increases in H₂O₂ content, activities of catalase and polyphenol oxidase, and malondialdehyde content. Application of dimethylthiourea (DMTU) and diphenylene iodonium (DPI) to the wounded and ATP-treated leaves, respectively, reduced the wound- and ATP-induced local and systemic increases in H₂O₂ content, activities of catalase and polyphenol oxidase, and malondialdehyde content. Moreover, the wound- and ATP-induced systemic increases of these physiological parameters were suppressed when DMTU or DPI applied to leaf petiole of the wounded and ATP-treated leaves. These results suggest that eATP at wounded sites could mediate the wound-induced local and systemic responses by H₂O₂-dependent signal transduction.     

Cell-surface-localized ATP detection with immobilized firefly luciferase

We demonstrate the application of an assay technique for the detection of ATP efflux from the cell surface. Until recently, the firefly luciferase assay has conventionally been used to detect ATP release because of its high sensitivity and wide detection ability. However, since this technique can be used only to infer the amount of diffused ATP in bulk solutions, it does not accurately reflect the dynamic ATP flux at the cell membrane. We therefore constructed luciferase fusion proteins that could be immobilized onto the cell surface. However, the low activities and limited application ranges of these proteins prompted us to use biotinylated luciferase given its attributes of strong and stable localization. Furthermore, luciferase can be immobilized strongly onto the biotinylated cell surface via streptavidin-biotin binding and can thus be used to determine the dynamic release of ATP near the cell surface.