Release of nicotinamide adenine dinucleotide ( β -NAD) upon stimulation of postganglionic nerve terminals in blood vessels and urinary bladder

1, -etheno-ATP; eADP, N 6 -etheno-ADP; 1, N 6 -etheno-AMP; 1, 6 -etheno-ADO; eADPR, N 6 -etheno-ADPR;


Introduction
Chemical signaling constitutes a major mechanism in neuroeffector transmission in the central and peripheral nervous systems. Postganglionic nerve terminals, in particular, characteristically release multiple factors upon action potential, a process referred to as plurichemical neurotransmission (1) or co-transmission (2). The evidence is particularly strong for postganglionic sympathetic nerves, which have been shown to co-release adenosine 5'triphosphate 1 (ATP), NE, and neuropeptide Y, factors that presently fulfill transmitter criteria (3,4,5,6). Postganglionic parasympathetic nerve terminals, on the other hand, release acetylcholine and ATP (2). In addition, neural control of effector cells involves molecules that implement potent excitatory or inhibitory actions at the neuroeffector junction in the absence of stringent evidence for transmitter function. These effects, collectively referred to as neuromodulation, are usually poorly understood, often due to lack of potent and selective antagonists (6); nonetheless, neuromodulation is an important mechanism, as is neurotransmission (7,8).
We report here that postganglionic nerve terminals in several smooth muscle preparations that are under autonomic nervous system control release novel factor(s) along with NE, ATP and the ATP metabolic products ADP, AMP, and ADO. We hypothesized that this novel factor has a nucleotide/nucleoside structure and acquires fluorescence properties upon etheno-derivatization.
Extensive screening revealed that the compound is a member of the β-NAD/cADPR/ADPR family. Further experimentation in canine mesenteric artery and vein, and canine and murine urinary bladders established that all three nucleotides, β-NAD, cADPR and ADPR, are present in tissue superfusates collected during EFS at parameters specific for neural activation, β-NAD being the prevailing nucleotide. Smaller amounts of β-NAD, cADPR and ADPR were found in 6 rubbed with rough-surface needle and perfused with distilled water to remove the endothelium. This procedure has been shown to successfully remove endothelium while smooth muscle contractility remains intact (9).

Overflow experiments:
The tissue segments were placed in 200-µl water-jacket BRANDEL superfusion chambers as described previously (10,11,12,13). Briefly, after 45 min equilibration, the tissues were subjected to a 15-s "conditioning" stimulation with a train of square wave pulses of 0.1 ms duration and a frequency of 4 Hz. Previous experiments have shown that the transmitter overflow evoked by stimulations subsequent to the conditioning stimulation is more consistent than in the absence of a conditioning stimulation. Thirty minutes after the conditioning stimulation the preparations were subjected to EFS for 60-120 s with a train of supra-threshold pulses of 0.1-0.3 ms at 4-16 Hz depending on the tissue. The EFS parameters were chosen after preliminary optimization experiments and verification that NE overflow is abolished by either tetrodotoxin (0.3-1 µmol/L) or ω-conotoxin GVIA (5 nmol/L). Samples of the superfusion solution were collected before the electrical stimulation (resting overflow) and during the electrical stimulation (electrically evoked overflow) in ice-cold test tubes. Samples were analyzed for nucleotide/nucleoside contents by HPLC techniques with fluorescence detection as described previously (12,13,14). Aliquots of some superfusate samples were processed for NE assay by HPLC techniques with electrochemical detection as described previously (15).
A more detailed characterization of the EFS-evoked overflow of nucleotides, nucleosides (i.e., ADO), and NE was carried out in canine isolated mesenteric arteries. In some experiments the tissue segments were superfused with either TTX (0.3-1 µmol/L) or ω-conotoxin GVIA (5 nmol/L) for 30 min before EFS to block neuronal fast Na + channels and N-type Ca 2+ channels, respectively. Other tissue segments were treated with guanethidine (3-10 µmol/L) for 1.5 h to inhibit action potential-induced activation of sympathetic nerve terminals. In some experiments tissues were treated with 6-OHDA using a method modified from Aprigliano and Hermsmeyer (16) to "denervate" adrenergic neurotransmission. Briefly, a modified Krebs-Ringer solution that lacked NaHCO 3 and NaH 2 PO 4 was prepared and the pH was adjusted to 4.9 using 20 µmol/L glutathione. 6-OHDA was dissolved in this solution at a concentration of 300 µg/ml. The tissue was placed in the 6-OHDA solution for 10 min. The tissue was then placed in regular oxygenated Krebs for 30 min, and then subsequently placed in the 6-OHDA solution for an additional 10 min. The tissue was then loaded in the chambers and equilibrated as usual. Samples were collected after 2 hr 15 min perfusion with regular Krebs. Parallel "no-6OHDA" timecontrols were carried out, as well. In some experiments NE content of the "pre-stimulation" and "during-stimulation" samples was also determined in the absence or presence of exogenous nucleotides (e.g., β-NAD, cADPR, ADPR, α-NAD, NAAD, NGD, cGDPR, 8-Br-cADPR, and ADO, 100 nmol/L). To test the possibility that contraction of the smooth muscle is the primary cause of nucleotide/nucleoside overflow, tissues were perfused either with ET-1 (0.05 µmol/L), U46619 (1 µmol/L), ANGII (0.1 µmol/L) or PMA (0.1 µmol/L) for 20 min.

Sample Preparation
A method modified from Levitt et al. (17), which originally describes a procedure for detection of 1,N 6 -etheno-derivatives of ATP, ADP, AMP and ADO, was employed. Briefly, 100 µl of a citrate phosphate buffer (pH 4.0) was added to 200 µl of the superfusate sample in a borosilicate glass culture tube (Fisher Scientific, USA). Chloroacetaldehyde was synthesized 8 according to a method modified from Secrist et al. (18) and Levitt et al. (17) and described previously (14). Ten µl of 2-chloroacetaldehyde was added to the samples in a fume hood; the culture tubes were covered with glass marbles, and the samples were heated for 40 min at 80 o C in a dry bath incubator (Fisher Scientific, USA) to produce 1,N 6 -etheno-nucleotides and 1,N 6etheno-nucleosides.
In some experiments authentic nucleotides were subjected to etheno-derivatization with 2-chloroacetaldehyde at 80 o C and the samples were desalted as described previously (14). The nucleotides were then eluted with a binary system consisting of eluent A (water) and eluent B (100 % acetonitrile) by a gradient according to the following linear program: time 0, 0 % eluent B; 40 min, 80 % eluent B; 80 min, 80 % eluent B. The flow rate was 0.8 ml/min. The absorbing fraction was collected. The fraction was then lyophilized and kept at -16 o C until processed for mass spectra verification of the etheno-nucleotide.

HPLC assay of etheno-nucleotides and etheno-nucleosides
The liquid chromatographic system used throughout this study was an HP1100 LC module system (Agilent Technologies, Wilmington, DE) as described previously (14). The mobile phase comprised 0.1 mol/L KH 2 PO 4 (pH 6.0) as eluent A; eluent B consisted of 35 % methanol and 65 % eluent A. Gradient elution was employed according to the following linear program: time 0, 0 % eluent B; 18 min, 100 % eluent B. Flow rate was 1 ml/min, run time 20 min and post-run time 5 min. Column temperature was ambient while the autosampler temperature was 4 o C. The fluorescent detector was set to record signals at an excitation wavelength of 230 nm and emission wavelength of 420 nm, which are the optimum conditions for detection of etheno-derivatives of nucleotides and nucleosides as shown previously (14). The non-derivatized compounds were detected at an excitation wavelength of 270 nm and emission wavelength of 410 nm according to preliminary optimization of the HPLC application.

Fraction Collection and Sample Concentration
To identify the compound that is released during EFS in canine mesenteric arteries, superfusate samples from 12 chambers containing ~65 mg tissue per chamber were placed together in two 2-ml Eppendorf tubes containing the pre-stimulation samples and the samples collected during EFS, respectively. The two resulting samples were further concentrated by Speed Vacuum (Savant SVC100, Thermo Electron Corp., Westmont, IL) to 1 ml volume. Seven hundred and fifty µl of each concentrated sample were injected into the HPLC system and 400 µl-fractions corresponding to the retention times of cADPR (7.0-7.4 min, "7.2-min fraction"), ADPR (8.3-8.7 min, "8.5-min fraction"), and β-NAD (10.3-10.7 min, "10.5-min fraction") were collected in borosilicate culture tubes containing 160 µl citric buffer. The exact retention times for the three nucleotides were determined by injecting β-NAD, cADPR and ADPR standards (40 nmol/injection) in the same sequence prior to the concentrated superfusate samples. The fractions were further subjected to etheno-derivatization with 17 µl 2-chloroacetaldehyde as described in HPLC Assay of Etheno-nucleotides and Etheno-nucleosides. The derivatized samples were injected into the HPLC and analyzed for 1,N 6 -etheno-ADPR content. To identify the compound that is released in urinary bladder, superfusate samples from 16-20 chambers containing either canine detrusor strips (~ 70 mg/chamber) or mouse bladder detrusors (2 bladders averaging ~15 mg per chamber) were processed as described for canine mesenteric artery segments. The experiments were performed in triplicate with each species.
The HPLC systems were controlled, and data collected, by a HP Kayak XA computer equipped with HP ChemStation (A.06.03) software from Agilent Technologies (Wilmington, DE, USA).
The amounts of NE in each sample were calculated from calibration curves of NE standards run simultaneously with every set of unknown samples. Results were normalized for sample volume and tissue weight and the overflow of NE was expressed in fmol/mg tissue.

MALDI-MS identification
The chemical identities of the molecules in the absorbing fraction from reversed-phase HPLC-FLD of ADPR standards, β-NAD standards, and cADPR standards subjected to ethenoderivatization were validated by matrix-assisted laser desorption/ionization mass spectrometry

Degradation of β-NAD and cADPR in contact with tissue
Canine mesenteric artery segments were loaded in a BRANDEL superfusion system as described in Overflow Experiments. Following a 45-min equilibration period the tissues were superfused either with β-NAD (0.2 mmol/L), cADPR (0.2 mmol/L) or NGD (0.2 mmol/L). A 200-µl sample from the beaker containing the substrate (no tissue present) was collected (S1).
The superfusion was stopped and 2 min later the content of the chamber containing the tissue was drained and collected (S2). All samples were collected in ice-cold test tubes and the reaction was stopped with liquid N 2 . The samples were then processed for nucleotide detection by a reverse phase HPLC technique in conjunction with fluorescence detection as described above.
Ecto-enzyme activity was determined by measuring the amount of substrate that had decreased in the S2 as compared to S1.

Force development
Ring preparations (5 mm long) of canine isolated mesenteric artery were mounted in 3 ml organ baths by inserting two stainless steel triangle mounts into the lumen and force displacements were further investigated as described previously (9). A resting force of 1 g was applied to the arterial segment. This was found to stretch vessels to near the optimum length for

Statistics
Data are presented as means ± s. e. mean. Means were compared by analysis of variance (one-way and two-way ANOVA) (GraphPadPrism v. 3, GraphPad Software, Inc.). A probability value of less than 0.05 was considered significant. 2B) smooth muscle preparations upon EFS, including canine isolated mesenteric vein, guineapig isolated mesenteric artery and vein, rat and murine tail artery, canine, guinea-pig, rabbit and murine urinary bladder, and rat vas deferens. We have also detected this peak in chromatograms from tissue superfusate samples of stimulated rat tail artery, rat and guinea-pig vas deferens and monkey urinary bladder (unpublished observations). Therefore, release of this factor is not restricted to one species or one nerve-smooth muscle preparation and may have a universal role.

EFS evokes overflow of nucleotides, nucleosides, and NE
The compound that elutes at ~11.2 min is subjected to changes by agents affecting neurotransmitter release (see below) and, therefore, might represent a novel factor in the autonomic nervous system control. The present study was designed, therefore, to identify this novel factor or the chemical group to which this compound belongs.
Given the fact that the 11.2-min peak can only be observed after etheno-derivatization of the superfusate samples, we hypothesized that this compound has a nucleotide/nucleoside structure.

Screening of 1,N 6 -etheno-derivatives of nucleotides and nucleosides
The general approach was to subject authentic compounds with nucleotide or nucleoside structures, for which important intracellular or extracellular functions are well-established, to etheno-derivatization according to an identical procedure as for tissue superfusates (discussed in Methods). Thus, uridine, guanidine and adenine nucleotides or nucleosides, as well as some compounds with di-nucleotide strictures, were tested. Table 1 shows the retention times of the compounds after subjecting various nucleotides, nucleotsides and dinucleotides to ethenoderivatization. Neither UTP nor guanidine derivatives (i.e., GTP, GDP, GMP, cyclic GMP), nor diguanosine polyphosphates (Gp n G) eluted at ~11.2 as the peak of interest ( β-NAD and its metabolic products cADPR and ADPR were also considered potential candidates, taking into account their nucleotide structure, potent intracellular and potential extracellular roles (19,20). While the elution time of the authentic ADPR was ~8.2 min (Fig.   3A), the derivatized product of ADPR (presumably etheno-ADPR) eluted at ~11.2 min (Fig. 3B), suggesting that the compound from the tissue superfusates having undergone derivatization with 2-chroloacetaldehyde is likely to be 1,N 6 -etheno-ADPR. Indeed, MALDI-MS analysis of the converted product showed a major positive ion of 583 mass units, which is 24 mass units higher than the 559 mass units measured for the positive ion of ADPR (Fig. 3C, D). The mass difference is consistent with 1,N 6 -etheno-ADPR being larger than ADPR and confirms that the experimental procedure of etheno-derivatization indeed produces 1,N 6 -etheno-derivatives from the authentic nucleotide compounds.
The compound was separated by reverse phase HPLC and detected as one peak at 11.2±0.2 min.
The retention time is identical to one, which corresponds to 1,N 6 -etheno-ADPR (as seen in Fig.   3B). The structure of the nucleotide produced by derivatization of cADPR at 80 o C has been determined by MALDI-MS as 1,N 6 -etheno-ADPR with a major positive ion of 583 mass units.
After derivatization of β-NAD at 80 o C and pH 4, two peaks were observed: the major peak eluted at ~11.2 min and a smaller peak eluted at 12.8±0.2 min. On the basis of the elution times of standard 1,N 6 -etheno-NAD (Sigma) and 1,N 6 -etheno-ADPR (Fig. 3B) the first peak was identified as 1,N 6 -etheno-ADPR and the second peak was identified as 1,N 6 -etheno-NAD. When derivatization was carried out at room temperature for 48 h, the major product of β-NAD was 1,N 6 -etheno-NAD. The molecular masses of both products were also determined by mass

Identification of released nucleotide in tissue superfusates
We then carried out experiments to identify which one, β-NAD, cADPR or ADPR, is the nucleotide released upon EFS in the canine mesenteric artery and vein, canine urinary bladder, and murine urinary bladder. We based our approach on the finding that each of the three nucleotides has a different elution time, but after reacting with 2-chloroacetaldehyde at 80 o C, pH 4.0 for 40 min, each nucleotide forms 1,N 6 -etheno-ADPR that elutes at ~11.2 min. Thus, if an HPLC fraction of the tissue superfusate does not produce a peak at 11.2 min after reacting with 2-chloroacetaldehyde at 80 o C for 40 min, then this fraction contains neither β-NAD, nor cADPR or ADPR. In contrast, if a fraction collected at a retention time specific for either β-NAD, cADPR or ADPR produces 1,N 6 -etheno-ADPR (and hence a peak at ~11.2 min) after ethenoderivatization then this fraction contains the corresponding nucleotide. As described in Methods, superfusate samples collected before (PS) and during EFS (ST) of 12-20 perfusion chambers were combined, concentrated and injected in the HPLC system. Fractions corresponding to the retention times of cADPR (7.2-min fraction), ADPR (8.5-min fraction) and β-NAD (10.5-min fraction) were then collected, etheno-derivatized and re-injected in the HPLC system. Fig.   6A,B,C (left column of the figure) show representative chromatograms from an experiment with canine mesenteric artery. In the 7.2-min fraction of the PS sample a peak at ~11.2 min was seen ( Fig. 6A, upper panel), suggesting that cADPR was present in the tissue superfusate at resting (no EFS) conditions. A peak of slightly greater size was seen in the samples collected during EFS (Fig. 6A, bottom panel).
Representative chromatograms are shown in Supplemental materials - Fig. 4. These results suggest that at least part of cADPR and ADPR that are detected in the tissue superfusates may be formed from β-NAD that is released upon EFS.

Contraction of the canine isolated mesenteric artery does not evoke overflow of β-NAD, cADPR or ADPR
To test the hypothesis that β-NAD/cADPR/ADPR are released upon contraction of the smooth muscle, we applied several agents that induce either receptor-mediated-(i.e., ET-1,

EFS-evoked overflow of ADPR in canine isolated mesenteric artery depends on stimulation frequency
The content of both 1,N 6 -etheno-ADPR (Fig. 7A) and NE (Fig. 7B) is reduced by chemical denervation with either TTX or guanethidine, or after disrupting the adrenergic nerve terminals with 6-OHDA. These findings taken together suggest that the release of β-NAD depends on the degree of neural activity. Finally, ω-conotoxin GVIA, a specific and selective Ntype Ca 2+ channel blocker (21), abolished the release of both β-NAD and NE (Fig. 7A,B) suggesting that Ca 2+ entry via N-type voltage-operated Ca 2+ channels is necessary for the release of both NE and β-NAD.

Exogenous β-NAD, cADPR and ADPR modulate the release of NE
To test the hypothesis that ADPR and the other members of the β-NAD/cADPR/ADPR axis modulate neurotransmitter release, we studied the effects of exogenously applied nucleotides on the EFS-evoked release of NE in canine mesenteric artery. The EFS (16Hz)-
Once released, ATP is rapidly converted to ADP, AMP and ADO by membrane-bound (23,24,25) and perhaps soluble or releasable nucleotidases (26,27). The amounts of these nucleotides and ADO in superfusate samples from small tissue preparations are usually below the threshold of most detection methods. One particularly useful approach of attaining detectable ranges of these endogenous compounds is their chemical conversion to 1,N 6 -etheno-derivatives (10,12,13,14,17). This maneuver increases approximately 1,000,000 fold the fluorescence coefficient of tested nucleotides and nucleosides. The present study confirms previous studies of ours and others that upon short-duration pulse stimulation of postganglionic nerve terminals, a cocktail containing ATP, ADP, AMP, ADO, and NE is released (10,11,12,13). The factors are present in different proportions depending on tissue type, parameters of EFS, and kinetics of neurotransmitter removal; their release clearly depends on the frequency of EFS. The current investigation demonstrates the presence of additional compounds in the tissue superfusates. The overflow (release) of these factors is associated with the degree of neural activity. They appear to be released in numerous vascular and non-vascular neuromuscular preparations and implicate involvement of novel mechanisms in the autonomic nervous system control. The present study was aimed at chemical identification and initial functional characterization of these factors. The general strategy was to identify: (i) the chemical structure; (ii) possible source, and (iii) potential role of this factor(s) at the neuromuscular junction in vascular and non-vascular smooth muscles.
We assumed a nucleotide/nucleoside structure for this compound since it can only be Likewise, etheno-derivatization at 80 o C of the two previously described derivatives of β-NAD, cADPR and ADPR (20,28) also produced fluorescent signals with the same elution time of 11.2 min. MALDI-MS analysis revealed that 1,N 6 -etheno-ADPR is the major product after ethenoderivatization at 80 o C of all three: β-NAD, cADPR and ADPR. It appears, therefore, that both β-NAD and cADPR undergo initial chemical conversion to ADPR, which is then derivatized to 1,N 6 -etheno-ADPR. This is in agreement with previous reports that cADPR undergoes spontaneous hydrolysis at acidic conditions (20) and that this process is accelerated at higher temperature (29). In the present study we also verified that at 80 o C and acidic conditions both β-NAD and cADPR are quickly hydrolyzed to ADPR, which then in the presence of 2chloroacetaldehyde is converted to 1,N 6 -etheno-ADPR. At our routine reaction conditions, therefore, we cannot distinguish whether one or more nucleotides from the β-NAD/cADPR/ADPR family are released upon EFS, before being chemically converted in vitro into the detectable fluorescent product 1,N 6 -etheno-ADPR. Methods utilized to measure endogenous concentrations of cADPR and β-NAD in cell extracts including radioimmuno (30,31), HPLC-UV (29) and cycling (32) assays cannot be utilized for detecting the nucleotides in the tissue superfusates investigated in the present study because these methods: 1) do not always discriminate between cADPR and ADPR, and 2) have insufficient sensitivity (i.e., nanomolar range) for detecting endogenous nucleotides in small tissue superfusates. Instead, our approach to identify the nucleotide that is released in tissue superfusates was based on the following reasoning: (i) the sensitivity of the FLD detection of nucleotides is increased when nucleotides are subjected to etheno-derivatization, so that small amounts of released nucleotides that usually go undetected (as is the case with neurotransmitters) could be measured, (ii) authentic β-NAD, β-NAD is converted into ADPR by NAD glycohydrolase and into cADPR by ADPribosyl cyclase, whereas cADPR can further be degraded to ADPR by cADPR hydrolase (28). Therefore, at least part of cADPR and ADPR that are present in tissue superfusates might be produced from the released β-NAD. In the present study we found that the concentrations of exogenous β-NAD, cADPR, as well as the ADP-ribosyl cyclase substrate NGD, are reduced in contact with artery segments, suggesting that indeed cADPR and ADPR can be produced from β-NAD in this system. It also appears that β-NAD is constitutively released since this compound together with small amounts of cADPR and ADPR are also present in tissue superfusates collected in the absence of nerve stimulation.
Next, we were interested to find whence β-NAD might originate. Our approach was to apply analogies to well-known factors at the neuromuscular junction such as NE and ATP. It is generally accepted that NE originates exclusively from sympathetic neurons. ATP, however, may originate from either neuronal or extraneuronal sources. The endothelium is considered to be the major extraneuronal source of ATP (32,33). However, the present study employed endothelium-denuded vessels ruling out this cell type as a major source for β-NAD under our experimental conditions. Besides the endothelium, it has been suggested that ATP can originate from the smooth muscle cells upon contraction (34). In the present study, however, the content of 1,N 6 -etheno-ADPR in tissue superfusates was not increased during contraction of the vessels with several contractile agents including the most potent vasoconstrictor yet described, ET-1. We conclude, therefore, that the contraction of the vascular smooth muscle per se does not cause release of neither β-NAD, cADPR nor ADPR in the canine mesenteric artery.
In canine isolated mesenteric artery preparations, the amount of 1,N 6 -etheno-ADPR in samples collected during EFS at 16 Hz exceeded the amount of 1,N 6 -etheno-ADPR in samples collected during 4 Hz, as did the EFS-evoked overflow of NE. In addition, the content of 1,N 6etheno-ADPR was reduced by chemical denervation with TTX, guanethidine, or 6-OHDA indicating that the release of β-NAD was action potential-induced. The release of NE was inhibited upon action of these agents, as well. Finally, ω-conotoxin GVIA, a specific and selective blocker of neuronal N-type Ca 2+ channels, abolished the release of both β-NAD and NE. Thus, release of β-NAD depends on the degree of neural activity, on the influx of extracellular Ca 2+ via N-type Ca 2+ channels, and generally parallels the release of NE and may, therefore, originate from sympathetic nerve terminals. The release of β-NAD in non-vascular preparations (i.e., murine urinary bladder) also depends on neural activity (our unpublished observations). Further studies are needed to resolve the exact origin and detailed characteristics of β-NAD release; however, it is clear that the release of this nucleotide shows a steep correlation with the activity of peripheral nerve terminals in various smooth muscle preparations.
We then asked what the role of these nucleotides might be at the vascular neuromuscular junction. The β-NAD/cADPR/ADPR system has well known intracellular functions. For example, β-NAD serves as a coenzyme for cellular oxidation-reduction reactions and a precursor of ADPR in the posttranslational modification of proteins, cADPR shows potent Ca 2+ -releasing activity from ryanodine-sensitive stores in a wide variety of cells, and ADPR modulates the function of membrane ion channels in addition to serving a well-known role in posttranslational modification of proteins (reviewed in 28,35). This system, therefore, appears to be involved in the regulation of cell functions related to intracellular Ca 2+ handling and membrane ion channel activity and, hence, might be an important factor in the smooth muscle neuroeffector process.
Besides the aforementioned functions that β-NAD, cADPR and ADPR play as important intracellular constituents, it is likely that these molecules also play extracellular roles.
Interestingly, in the present study exogenous application of β-NAD, cADPR and ADPR (100 µmol/L) reduced the release of NE. At present it is not clear whether the inhibitory effect of β-NAD we observed is mediated by β-NAD itself or by its metabolic products ADPR and cADPR.
It is also unclear whether these effects are mediated by other nucleotide metabolites such as ATP and ADO. In the present study we showed that other nucleotides with similar structures including α-NAD, NAAD, NGD, cGDPR, and 8-Br-cADPR (100 µmol/L) did not significantly reduce the release of NE. Interestingly, exogenous ADO, which is a major metabolite of β-NAD, cADPR and ADPR, even facilitated the release of NE in the canine mesenteric artery possibly via facilitatory adenosine A2 receptors as shown in other systems (36,37). Although the mechanisms of action are far from clear, the present study suggests that the β-NAD/cADPR/ADPR axis may have a novel role in neuromodulation. Further work will be necessary to resolve the full range of extracellular functions and the mechanisms of extracellular action of β-NAD and its metabolites at the smooth muscle neuroeffector junction.
In conclusion, we have identified constitutive and nerve-evoked release of β-NAD and its metabolic products cADPR and ADPR in numerous smooth muscle preparations. These nucleotides appear to regulate the release of NE at the vascular neuroeffector junction.
Therefore, the β-NAD/cADPR/ADPR system constitutes a novel pathway in the autonomic nervous system control.             In all cases the substrate concentration was reduced during the 2-min contact with the artery strips. In the case of NGD as substrate an increase of the product cGDPR is also observed (B).