Abstract
Recently, we have shown that by releasing specific nucleotidases the sympathetic nerves of the guinea pig vas deferens may regulate the metabolism of extracellular adenine nucleotides and consequently, the inactivation of neurotransmitter ATP. Based on the evidence for tetrodotoxin sensitivity and calcium dependence of the nerve stimulation-evoked overflow of enzyme activity, we have suggested that soluble nucleotidases may be stored in synaptic vesicles within the sympathetic nerves and released upon arrival of nerve action potentials by a mechanism similar to that for release of neurotransmitters. To further test this hypothesis we studied the time course of nerve stimulation-evoked overflow of ATP, norepinephrine (NE), releasable ATPase (r-ATPase) activity, and releasable AMPase (r-AMPase) activity under control conditions and in the presence of drugs known to selectively modulate sympathetic neurotransmission. The results show that the time course of overflow of r-ATPase and r-AMPase activities resembles the transient pattern of overflow of ATP but not the tonic pattern of overflow of NE. Vasa deferentia dissected from animals treated with reserpine release ATP, r-ATPase, and r-AMPase, whereas the overflow of NE is completely abolished. Guanethidine, on the other hand, inhibits equally well the overflow of the two neurotransmitters and the releasable nucleotidase activities. Agonists of the α2-adrenergic receptors abolish the overflow of ATP, r- ATPase, and r-AMPase but not the overflow of NE. This evidence supports the idea that the sympathetic nerves of the guinea pig vas deferens store and release ATP together with specific nucleotidases responsible for the inactivation of this neurotransmitter.
Previously, we have demonstrated that stimulation of the sympathetic nerves innervating the guinea pig vas deferens greatly accelerates the breakdown of extracellular ATP (Todorov et al., 1996). This nerve stimulation-related metabolism of ATP appears to be associated with a release of nucleotidase activity, which overflows the tissue preparations and can be detected in samples of the superfusing solution in soluble form (Todorov et al., 1997).
The releasable nucleotidase activity exhibits some features that are common for both ecto-ATPases and for ecto-5′-nucleotidases. Like the ectonucleoside triphosphate diphosphohydrolases (E-NTPDases) (for review, see Zimmermann and Braun, 1999) the releasable enzymes breakdown ATP and ADP and this metabolism is prevented in the presence of suramin and ARL 67156 (Kennedy et al., 1997; Todorov et al., 1997;Westfall et al., 2000). Inhibitors of classical P-type, Fo-type, or V-type ATPases are not effective inhibitors of the releasable ATPase activity. Unlike E-NTPDases the releasable enzymes breakdown AMP to adenosine (ADO) and this metabolism is blocked by α,β-m ADP and concanavalin A, blockers of ecto-5′-nucleotidases, but not by suramin (Mihaylova-Todorova et al., 2000b). Based on these findings we have suggested that the releasable neuronal nucleotidase activity comprises two separate activities, an ecto-ATPDase (E- NTPDase)-like activity and an AMPase (ecto-5′-nucleotidase)-like activity that are differentially modulated by inhibitors. It is not known at this point whether these activities belong to a single protein or to separate proteins.
Both the breakdown of ATP and the overflow of enzyme activity were abolished when neurotransmission in the guinea pig vas deferens was inhibited by tetrodotoxin or cadmium (Cd2+), or under nominally Ca2+-free conditions (Todorov et al., 1997). This evidence, plus the findings that PC12 cells (S. Mihaylova-Todorova, unpublished observation) also release nucleotidases, lead us to speculate that nucleotidases may be stored in synaptic vesicles within the sympathetic nerves and released upon arrival of nerve action potentials by calcium-dependent exocytosis.
Results from chronopharmacological experiments have shown that in the guinea pig vas deferens the release of sympathetic cotransmitters involves at least two different mechanisms, one that releases ATP in a transient manner, and another that releases predominately NE in a sustained manner (Todorov et al., 1994, 1996). The release of the two cotransmitters appears to be controlled by Ca2+influx via different subtypes of voltage-activated Ca2+-channels, and modulated by different subsets of prejunctional α2-adrenergic receptors (Todorov et al., 1995; Westfall et al., 1996). Here we use the existing differences between the mechanisms for release of ATP and NE to test the hypothesis that the sympathetic nerves of the guinea pig vas deferens may store and release one of the cotransmitters together with one or both of the releasable enzymes involved in the metabolism of extracellular adenine nucleotides. The results from this study favor the hypothesis that the sympathetic nerves innervating the guinea pig vas deferens store and release soluble ATPase and soluble AMPase together with neurotransmitter ATP but not with its cotransmitter NE.
Materials and Methods
Tissue Preparation and Nerve Stimulation Protocol.
Male albino guinea pigs (350–400 g) were sacrificed by decapitation. The vasa deferentia were removed, cleaned of connective tissue, and the lumen exposed by section along the longitudinal axis. Three tissues, each from a different animal, were loaded into a 0.2-ml Brandel (Gaithersburg, MD) chamber. Whatman 541 filters were cut to fit both ends of the chamber, which was then inserted vertically into a thermostatic (36°C) water-jacketed block (Brandel) and closed with platinum screen electrodes at each end. The tissues were superfused at 2 ml/min from bottom to top with modified Krebs-HEPES buffer (pH 7.4) of the following composition: 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.5 mM CaCl2, 5 mM HEPES, and 11 mM glucose that was constantly bubbled with O2.
The tissue preparations were stimulated once by electrical field stimulation (EFS) for 60 s at a frequency of 16 Hz with square-wave pulses of 0.2-ms duration and supramaximal voltage and samples of superfusate (∼340 μl) were collected at 10-s intervals before (S0), during (S1–6), and after (S7) the EFS. When used, drugs were added to the superfusing solution 20 min before the onset of the EFS, unless stated otherwise. In experiments with guanethidine, the drug (100 μM) was present in the Krebs-HEPES solution during dissection of the tissues and thereafter for a total of 90 min before the EFS. Reserpine was dissolved at 2.5 mg/ml in sterile 20% ascorbic acid and administered by intraperitoneal injection at a dose of 1 mg/kg, 24 h before the experiment.
Chronopharmacological Technique.
We have used aliquots from each 10-s collection to quantify the amount of concomitantly released endogenous ATP, ADP, AMP, ADO, NE, r-ATPase, and r-AMPase. This method enables us to monitor the release of the two sympathetic cotransmitters and to compare their temporal pattern of release with the temporal pattern of release of the nucleotidases.
HPLC Analysis of Endogenously Released ATP.
The amount of ATP released during sympathetic nerve stimulation was estimated from the sum total of the detected ATP and its degradation products (ADP, AMP, and ADO) present in the superfusate, as described previously (Todorov et al., 1996). Briefly, a 200-μl aliquot taken from each 10-s collection of superfusate was acidified with 92 μl of citric phosphate buffer, pH 4, and incubated for 40 min at 80°C in a dry heating block in the presence of 8 μl of 2-chloroacetaldehyde. During the incubation, ATP, ADP, AMP, and ADO present in the sample were transformed to their respective fluorescent derivatives 1N6-etheno-ATP (eATP), 1N6-etheno-ADP (eADP), 1N6-etheno-AMP (eAMP), and 1N6-etheno-ADO (eADO). Two hundred microliters of the mixture, containing the etheno-adenine nucleotides and etheno-adenosine, were injected by means of Waters 715 Ultra Wisp sample processor (Waters Co., Milford, MA) onto a 4-μm, 8 × 10-mm, Nova-Pack Phenyl cartridge (Waters Co.) connected to Waters 510 HPLC pump. Bound nucleotides were eluted at flow rate of 2 ml/min with gradient change from buffer A (0.1 M KH2PO4, adjusted to pH 6.0 with NaOH) to buffer B (25% methanol in buffer A) as follows: 1 to 8 min, from 100% A to 100% B; 8 to 10 min, 100% B; 10 to 13 min, from 100% B to 100% A; and 13 to 17 min 100% A. Fluorescence of eATP, eADP, eAMP, and eADO, at retention times 4.5 ± 0.2, 5.7 ± 0.2, 7.8 ± 0.2, and 12.5 ± 0.2 min, respectively, was quantified with an excitation wavelength of 230 nm and an emission wavelength of 420 nm by Shimadzu RF 535 fluorescent monitor (Columbia, MD).
HPLC Analysis of Endogenously Released NE.
To measure the overflow of NE, 80-μl aliquots from each 10-s collection of superfusate were acidified with 20 μl of 1 M perchloric acid and filtered through a 0.22-μm Cameo 3N syringe filter (Westborough, MA) into limited volume inserts (Waters Associates, Milford, MA) by centrifugation at 1000g for 1 min. Acidified sample (30 μl) was injected by Waters 715 Ultra Wisp sample processor (Waters Co.) onto a catecholamine HR-80 column (ESA, Chelmsford, MA), supported by Waters 510 HPLC pump. Catecholamines were eluted at isocratic regime using mobile phase: 50 mM Na2PO4, 0.2 mM EDTA, 3 mM 1-heptanesulfonic acid, 3% methanol (v/v) in deionized water, and pH 2.6 adjusted with phosphoric acid, at a flow rate of 1.5 ml/min. NE (retention time 2.9 ± 0.2 min) was quantified using dual electrode coulometric electrochemical conditioning cell model 5021 and Coulochem II detector (ESA).
The HPLC systems were controlled by, and data collected by, an HP Vectra XU computer equipped with an LAC/E card and Millennium 32 Chromatography Manager software from Waters Co. Identification of individual peaks in chromatograms was by comparison with the retention times of known amounts of respective etheno-adenine nucleotides, ADO, or catecholamines and the concentration was determined by peak area per picomole relationship compared with standards. Results were normalized for volume and tissue weight and the data calculated as picomoles per milligram per 10-s collection. When plotted versus time the results demonstrate the time course of overflow of the released endogenous NE as well as endogenous adenine nucleotide phosphates (ATP, ADP, AMP), ADO, or their sum total (PUR).
Preparation of 1N6-Etheno-Substrates.
Etheno-modified nucleotides have been previously used as substrates for ATPases (Secrist et al., 1972) or 5′-nucleotidases (Jamal et al., 1988;Bonitati et al., 1993). Here we use etheno-ATP or etheno-AMP as substrates of ATP-metabolizing or AMP-metabolizing enzymes, released in soluble form during nerve stimulation. Stock solutions of 1 mM eATP or eAMP were prepared from ATP or AMP, respectively, as described above and stored at −20°C until use.
Measurement of Nucleotide Hydrolysis by Releasable Nucleotidases.
To determine the nucleotidase activity in the overflow, 20-μl aliquots of each sample of superfusate collected at 10-s intervals were combined with 5 μl of 1 mM eATP (100 μM final concentration) in a final volume of 50 μl of Krebs-HEPES buffer, pH 7.4, and incubated for 60 min at 37°C. The reaction was stopped with the addition of 100 μl of ice-cold citric phosphate buffer (pH 4). The degradation products and remaining nonmetabolized substrate in a 100-μl sample were quantified by reverse phase chromatography and fluorescent detection as described above for assay of endogenous adenine nucleotides. The ATPase activity present in each sample was estimated from the sum of metabolites (eADP + eAMP + eADO, also called e-product) formed as percentage from initial substrate (eATP + eADP + eAMP + eADO). This value was further normalized for non-neuronal degradation present in the sample collected before the stimulation (S0). The resulting net product, plotted versus time, represents the time course of increase in hydrolyzing activity during EFS of the sympathetic nerves of the guinea pig vas deferens and reflects the release of soluble ATPase. The release of soluble AMPase was estimated from the rate of formation of eADO from initial substrate eAMP (100 μM) under similar conditions.
The effects of exogenous NE or clonidine on the catalytic activity of the releasable enzymes were tested in samples of superfusate collected at 10-s intervals (S0–7) as described for control experiments. NE (1, 10, and 100 μM) or clonidine (1 and 10 μM) was added to the reaction mixture 10 min before the addition of substrates.
Statistics and Graphic Presentation.
At least six replicates of the overflow experiments were performed in absence and in presence of each of the drugs. The normalized data are plotted as mean ± S.E.M., using GraphPad Software Prism 2.01.
Chemicals.
The following chemicals were purchased from Sigma (St. Louis, MO): adenosine 5′-triphosphate (disodium salt), adenosine 5′-diphosphate (sodium salt), adenosine 5′-monophosphate (sodium salt), adenosine (hemisulfate salt), chloroacetal, clonidine, guanethidine, HEPES, and EDTA. Methanol was from B&J (Muskegom, MI) and 1-heptanesulfonic acid was purchased from Fisher Scientific (Pittsburgh, PA). 2-Chloroacethaldehyde was prepared in the lab, as described previously (Todorov et al., 1996).
Results
There were no detectable amounts of NE, ATP, ADP, AMP, ADO, or ATPase or AMPase activity in the samples collected before the onset of the EFS. Stimulation of the sympathetic nerves of the guinea pig vas deferens caused the overflow of ATP, ADP, AMP, ADO, and NE, and the appearance of soluble ATPase and AMPase activities, presented, respectively, in plots A through D and F through H of Figs. 1 through3.
In control experiments (open circles in Figs. 1-3), the nerve stimulation-induced overflow of ATP, ADP, and AMP reached a peak by 20 s and then quickly declined to prestimulation levels even though the stimulation continued for 60 s. The overflow of ADO followed a similar pattern although it reached a peak by 30 s. Similarly, the sum total (purines) of ATP and its degradation products ADP, AMP, and ADO follows the transient time course of overflow of adenine nucleotides and adenosine (Figs. 1-3, E). The overflow of NE (Figs. 1-3, F), on the other hand, increased steadily until the end of the neurogenic stimulation.
A greater percentage of metabolites of eATP (Figs. 1-3, G) or eAMP (Figs. 1-3, H) was produced by the samples collected during the first three collection periods (S1–3) than after the 30th s of stimulation (S4–6). These results reflect a transient appearance of r-ATPase and r-AMPase activity in the beginning of the sympathetic nerve stimulation. The pattern of release of the ATPase and AMPase activities resembles the transient pattern of release of purines but not the sustained pattern of release of NE (Figs. 1-3, F).
In vitro exposure of vasa deferentia to the sympatholytic drug guanethidine (100 μM) (Fig. 1) substantially reduced the overflow of the neurotransmitters NE and ATP as well as the soluble ATPase and AMPase activities. The overflow of adenine nucleotides, adenosine, and the corresponding sum total of ATP and metabolites (PUR) was reduced by approximately 80% from their peak and this effect was associated with a loss of the typical transient time course of release (Fig. 1, A–E). The overflow of NE was also profoundly inhibited by guanethidine (Fig.1F). Likewise, the amplitude of both r-ATPase (Fig. 1G) and r-AMPase (Fig. 1H) activities was depressed by guanethidine in the degree seen for the neurotransmitters.
Overnight treatment of guinea pigs with reserpine (Fig.2) abolished the nerve stimulation-evoked overflow of NE (Fig. 2F) from the isolated vas deferens. Even though decreased in amplitude, the transient overflow of ATP, ADP, AMP, and ADO was not significantly affected by the reserpine treatment (Fig. 2, A–E). This same pattern applied for the overflow of both r-ATPase and r-AMPase activity (Fig. 2, G and H).
Treatment with the α2-adrenoceptor agonist clonidine (1 μM) (Fig. 3) produced only marginal inhibition of the NE overflow (Fig. 3F). The corresponding overflow of ATP, ADP, AMP, ADO (Fig. 3, A–D), and total purines (Fig.3E), however, was virtually abolished. The overflow of r-ATPase (Fig.3G) and r-AMPase (Fig. 3H) was also greatly inhibited by clonidine. Similar effects on the release of neurotransmitters and soluble nucleotidases were obtained with another α2-adrenoceptor agonist, xylazine (1 μM) (data not shown).
Exogenous NE (Fig. 4) in concentrations from 1 to 100 μM had no effect on the activity of r-ATPase (Fig. 4A) or r-AMPase (Fig. 4B). Similar results were obtained in the presence of 1 and 10 μM clonidine (data not shown).
Discussion
The experimental approach of this work is based on the assumption that the release of chemicals that are stored together in one synaptic vesicle should occur in parallel. In addition, drugs that modify the release of one of the chemicals should do likewise to the other.
The fact that tetrodotoxin, Cd2+, or the removal of Ca2+ blocks the metabolism of extracellular adenine nucleotides and prevents the appearance of nucleotidase activity in the superfusing solution prompted us to suggest that the sympathetic nerves themselves may release the enzymes that terminate the action of neurotransmitter ATP (Todorov et al., 1997). Here we demonstrate that guanethidine, a well known blocker of the adrenergic neurons (Ambache and Zar, 1971; Furness, 1974), abolishes the overflow of sympathetic cotransmitters ATP and NE as well as the overflow of both r-ATPase and r-AMPase activity. These results are consistent with the proposed neuronal origin of these soluble nucleotidases.
To determine the time course of release of the nucleotidases we have compared the enzyme activities of samples of superfusate collected in 10-s intervals during stimulation of the sympathetic nerves of the guinea pig vas deferens. The results demonstrate that changeable amounts of nucleotidase activity appear during 60 s of nerve stimulation. Both r-ATPase and r-AMPase reached a peak by about 20 s, and then quickly declined despite the continuing stimulation. The corresponding release of NE, however, follows a distinctively different time course: the overflow of NE reaches a peak later during the train of stimuli, and remains at a constant level until the end of stimulation. If NE acts as an inhibitor of the catalytic enzyme activity of the released nucleotidases, the higher concentrations of NE reached in the second part of the stimulation could suppress the activity of the nucleotidases, and this inhibition could account for the appearance of transient activation of otherwise steadily increasing release of nucleotidases. We excluded this possibility by showing that NE did not inhibit the ATPase or AMPase catalytic activity of samples of superfusate collected under control stimulation conditions. Thus, the transient time course of enzyme activities is most probably due to a transient release of the enzymes themselves. The release of nucleotidases clearly overlaps with the time course of concomitantly released endogenous ATP from this tissue. This result suggests that a different source and/or different mechanisms may be involved in the release of NE on one hand, and the release of ATP, r-ATPase, and r-AMPase on the other.
Overnight treatment of the guinea pigs with reserpine, an irreversible blocker of the synaptic vesicle monoamine transporter, causes depletion of peripheral sympathetic neurons of their NE content. Accordingly, in our experiments, the overflow of NE is abolished by reserpine. At the same time, the stimulated release of ATP and soluble nucleotidases is only marginally decreased but not abolished. It is also evident from the data that the breakdown of the released ATP is somewhat decreased. It seems that although the degradation from ATP to ADP occur normally, the next step, metabolism from ADP to AMP, is slightly inhibited and results in a build up of ADP. The reduction in the amounts of released nucleotidases may account for this effect. Analysis of the enzyme activity of the releasable enzymes has shown (Mihaylova-Todorova et al., 1998, 2000b) that the rate of the ADP degradation step is the slowest and most prone to further retardation by dilution of the enzyme mixture.
The amount of NE released from the sympathetic nerves of the guinea pig vas deferens appears to be negatively regulated by a subset of prejunctional α2-adrenoceptors that has little influence on the release of its cotransmitter, ATP (Driessen et al., 1993; Msghina et al., 1999; Todorov et al., 1999). Moreover, at high frequencies of stimulation (above 8 Hz), the negative feedback autoinhibition through α2-adrenergic receptors is expected to be fully engaged, due to saturating amounts of NE and further addition of exogenous α2-adrenoceptor agonists such as clonidine or xylazine is not expected to have significant additional inhibitory effect on the NE release. At the same time under these conditions, the release of neurotransmitter ATP is virtually abolished by clonidine or xylazine (Todorov et al., 1995; Westfall et al., 1996). Because clonidine does not suppress the catalytic r-ATPase or r-AMPase activity of releasable nucleotidases, here we use these discriminatory effects of the α2-adrenoceptor agonists to show that clonidine and xylazine inhibit preferentially not only the release of neurotransmitter ATP but also the nerve stimulation-evoked release of r-ATPase and r-AMPase.
Based on the evidence for similar temporal patterns and prejunctional modulation of release of ATP and neuronal nucleotidases, we hypothesize that ATP may be stored and released together with the enzymes responsible for its extracellular inactivation, presumably from the same synaptic vesicles within the sympathetic nerves.
It is not unusual for enzymes to be stored in an inactive form and to become activated upon arrival of a specific stimulus. Acidic pH, and/or lower calcium content of the intracellular storage compartment could be unfavorable factors for the activation of the hydrolytic activity and could permit costorage of enzymes with ATP. The evidence presented here does not exclude the possibility that ATP, or another factor released upon nerve stimulation, may promote release of the enzyme(s) from other sites. We conclude that the sympathetic nerves of the guinea pig vas deferens release neurotransmitter ATP together with the nucleotidases responsible for its inactivation.
Acknowledgment
The technical assistance of Jennifer Norman, Howard Hughes Foundation student trainee, is highly appreciated.
Footnotes
-
Send reprint requests to: Svetlana Mihaylova-Todorova, M.D., Ph.D., Department of Pharmacology, University of Nevada School of Medicine, Howard Medical Sciences Bldg. Rm. 222, MS 318, Reno, NV 89557-0046. E-mail: mihay_s{at}med.unr.edu
-
This work was supported by Grant HL 38126 from the National Institutes of Health. A version of this work has appeared as a proceeding article to the Second International Workshop on Ecto-ATPases and Related Ectonucleotidases, Diepenbeek, Belgium, June 14–18, 1999 (see Mihaylova-Todorova et al., 2000a).
- Abbreviations:
- E-NTPDase
- ectonucleoside triphosphate diphosphohydrolase
- ADO
- adenosine
- α,β-m ADP
- α,β-methylene 5′-adenosine diphosphate
- ATPDase
- adenosine tri-di-phosphatase, also apyraser-ATPase
- releasable ATPase
- r-ATPase
- NE
- norepinephrine
- r-AMPase
- releasable AMPase
- EFS
- electrical field stimulation
- eATP
- 1N6-etheno-ATP
- eADP
- 1N6-etheno-ADP
- eAMP
- 1N6-etheno-AMP
- eADO
- 1N6-etheno-ADO
- PUR
- sum total of ATP + ADP + AMP + ADO
- Received May 10, 2000.
- Accepted July 28, 2000.
- The American Society for Pharmacology and Experimental Therapeutics