β-Nicotinamide adenine dinucleotide acts at prejunctional adenosine A₁ receptors to suppress inhibitory musculomotor neurotransmission in guinea pig colon and human jejunum

Abstract

Intracellular microelectrodes were used to record neurogenic inhibitory junction potentials in the intestinal circular muscle coat. Electrical field stimulation was used to stimulate intramural neurons and evoke contraction of the smooth musculature. Exposure to β-nicotinamide adenine dinucleotide (β-NAD) did not alter smooth muscle membrane potential in guinea pig colon or human jejunum. ATP, ADP, β-NAD, and adenosine, as well as the purinergic P2Y₁ receptor antagonists MRS 2179 and MRS 2500 and the adenosine A₁ receptor agonist 2-chloro-N⁶-cyclopentyladenosine, each suppressed inhibitory junction potentials in guinea pig and human preparations. β-NAD suppressed contractile force of twitch-like contractions evoked by electrical field stimulation in guinea pig and human preparations. P2Y₁ receptor antagonists did not reverse this action. Stimulation of adenosine A₁ receptors with 2-chloro-N⁶-cyclopentyladenosine suppressed the force of twitch contractions evoked by electrical field stimulation in like manner to the action of β-NAD. Blockade of adenosine A₁ receptors with 8-cyclopentyl-1,3-dipropylxanthine suppressed the inhibitory action of β-NAD on the force of electrically evoked contractions. The results do not support an inhibitory neurotransmitter role for β-NAD at intestinal neuromuscular junctions. The data suggest that β-NAD is a ligand for the adenosine A₁ receptor subtype expressed by neurons in the enteric nervous system. The influence of β-NAD on intestinal motility emerges from adenosine A₁ receptor-mediated suppression of neurotransmitter release at inhibitory neuromuscular junctions.

firing of inhibitory musculomotor neurons in the enteric nervous system (ENS) accounts for relaxation of contractile tension in gastrointestinal (GI) smooth muscles (59). ATP was implicated as an inhibitory neurotransmitter at neuromuscular junctions in the GI tract in the early 1960s (4–6). Neurogenic inhibition, evoked by electrical field stimulation of the ENS or exposure to nicotinic receptor agonists in preparations from multiple species in vitro, was found to be suppressed by purinergic receptor antagonists, consistent with release of ATP as an inhibitory neurotransmitter at neuromuscular junctions in GI smooth muscles (2). Electrical field stimulation of preparations from human and nonhuman primate colon in vitro evoked neurogenic release of ATP directly into the superfusate, also consistent with a role for ATP as a neurotransmitter (33).

Inhibitory junction potentials (IJPs) reflect inhibitory input from the ENS to GI smooth muscles and are convenient markers in pharmacological study of neurotransmitter and receptor identification, as was the case in the present study. Transmural electrical field stimulation or focal electrical stimulation of interganglionic fiber tracts in the myenteric plexus of guinea pig, mouse, or human preparations evokes characteristic biphasic IJPs comprising a rapidly activating (“fast”) component followed by a slowly activating, longer-lasting (“slow”) component (23, 56, 65). The slow IJP is abolished by treatments that block the synthesis of nitric oxide and is mimicked by exogenous application of nitric oxide. The fast IJP is blocked by selective purinergic P2Y₁ receptor antagonists and mimicked by application of ATP. In general, according to Bennett (2), fast IJPs are mediated by ATP and, perhaps, pituitary adenylate cyclase-activating polypeptide, whereas slow IJPs involve motor neuronal release of nitric oxide and vasoactive intestinal polypeptide. Multiple lines of evidence continue to support the idea that ATP is the inhibitory neurotransmitter released by ENS musculomotor neurons. Nevertheless, other purines have been put forward as mediators of GI inhibitory neuromuscular transmission. β-Nicotinamide adenine dinucleotide (β-NAD) and its bioactive metabolite ADP-ribose (ADPR), rather than ATP, were proposed as the most probable purinergic neurotransmitters (18, 43, 50). However, the validity of the β-NAD-ADPR hypothesis has been questioned (29).

Introduction of the selective and potent competitive P2Y₁ receptor antagonist MRS 2179 was a significant step forward in identification of the P2Y₁ receptor as a target for ATP as an inhibitory neurotransmitter at smooth muscle junctions in the GI tract (45, 63). Results from pharmacological studies with MRS 2179 and other selective P2Y₁ receptor antagonists provide firm evidence for expression of P2Y₁ receptors by GI smooth muscles and support the idea that ATP is a ligand for the receptor and a mediator of IJPs (22, 25, 56). Furthermore, the finding that purinergic neurotransmission was absent in P2Y₁ receptor-knockout mouse colon, gastric antrum, and cecum added support for an ATP-P2Y₁ hypothesis (24, 28). On the other hand, neurogenic release of uridine adenosine tetraphosphate was recently reported to relax tension and hyperpolarize the membrane potential in smooth muscle of human and mouse colon in a manner sensitive to blockade by the P2Y₁ receptor antagonist MRS 2500 (19).

Hydrolysis of ATP, by ectonucleotidases, terminates its action at purinergic postjunctional receptors expressed by GI smooth muscles, as well as at postsynaptic purinergic receptors on ENS neurons. Ectonucleotidase activity converts ATP to ADP and AMP and AMP to adenosine. Each of these purines is implicated as a neurotransmitter or neuromodulator in the ENS (11, 12, 27, 31, 64). In view of the signaling roles of purines other than ATP, Hwang et al. (32) suggested that β-NAD might be released from sources within ENS ganglia, as well as at inhibitory motor nerve junctions in GI smooth muscles.

Based on reports that β-NAD might be an inhibitory neurotransmitter at motor nerve junctions in smooth muscles of the GI tract and that it is released inside ganglia of the ENS, we investigated if it might have a role as a neurotransmitter at synapses in the ENS or at prejunctional nerve terminals in the musculature (43). We obtained evidence for an action of β-NAD at transmitter release sites on ENS inhibitory musculomotor neurons. We were unable to confirm a role for β-NAD as an inhibitory neurotransmitter at neuromuscular junctions in the circular muscle coat of human or guinea pig intestine.

MATERIALS AND METHODS

Tissue preparation.

Methods for procurement of tissues for the in vitro studies are described elsewhere (53, 57). Male Hartley-strain guinea pigs (300–400 g body wt) were obtained from Charles River (Wilmington, MA). Animal care and experimental protocols were approved by The Ohio State University Laboratory Animal Care and Use Committee and followed the National Institutes of Health “Guide for the Care and Use of Laboratory Animals.” Segments of small intestine and colon for muscle bath studies and for electrophysiology were obtained by microdissection from guinea pigs. Fresh preparations of healthy human small intestine were obtained from segments of jejunum discarded during Roux-en-Y gastric bypass surgeries, as described elsewhere (53, 57). The human protocols were approved by the Institutional Review Board of The Ohio State University Office of Research Risks Protection (protocol 02H0208).

Electrophysiology.

Conventional methods for intracellular electrophysiological recording, with “sharp” glass microelectrodes in the circular muscle coat of intestinal preparations, were essentially as described elsewhere (56, 62). The microelectrodes were connected through preamplifiers (model 767, World Precision Instruments) equipped with negative-capacity compensation and bridge circuitry for injection of current through the recording electrode. Electrophysiological data were digitized, stored on disk, and analyzed with the PowerLab 5 data acquisition system/LabChart recorder. Neurogenic junction potentials in the circular muscle coat were evoked by focal electrical stimulation of interganglionic fiber tracts in the myenteric plexus. Pharmacological agents were applied in the bathing solution. Composition of the Krebs solution in all studies was as follows (in mM): 120 NaCl, 6 KCl, 2.5 CaCl₂, 1.2 MgCl₂, 1.35 NaH₂PO₄, 14.4 NaHCO₃, and 11.5 glucose. The Krebs solution in the Ussing chambers was bubbled with 95% O₂-5% CO₂ and buffered at pH 7.4.

Muscle contractile behavior.

Contractile behavior of strips of intestinal muscle or intact intestinal segments was recorded as we described previously (58). Routine organ bath pharmacological methods were used to analyze actions of purinergic receptor agonists and antagonists on contractile behavior of the strips or segments. Studies were done simultaneously in four tissue baths. The preparations were attached to a tungsten stimulating electrode used for electrical field stimulation. Analog changes in muscle contractile tension were digitized and stored on hard drives for analysis.

Agents.

ATP, ADP, ADPR, adenosine, 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), β-NAD, and 2-chloro-N⁶-cyclopentyladenosine (CCPA) were obtained from Sigma (St. Louis, MO). MRS 2179, MRS 2500, MRS 2365, TNP-ATP triethylammonium salt, endomorphin-2, and tetrodotoxin (TTX) were purchased from Tocris Bioscience (R & D Systems, Minneapolis, MN).

Data analysis.

Data were analyzed essentially as described elsewhere (36, 58). Values are means ± SE; n refers to the number of neurons. Concentration-response relationships were constructed using the following least-squares fitting routine: V = V_max/[1 + (EC₅₀/C)^nH], where V is the observed response, EC₅₀ is the concentration that induces the half-maximal response, and nH is the apparent Hill coefficient. Results from all experiments were averaged and fit to a single concentration-response curve by using Sigma Plot software (SPSS, Chicago, IL). Paired or unpaired t-test was used to determine statistical significance. P < 0.05 was considered statistically significant.

RESULTS

Smooth muscle membrane potential.

Bath application of ATP or MRS 2365, a selective agonist with an EC₅₀ of 0.4 nM for the P2Y₁ receptor, evoked hyperpolarization of the membrane potential in impalements of 36 circular muscle fibers in 24 preparations from the midcolon of 24 guinea pigs (Figs. 1A and 2) (8). Application of β-NAD, in micromolar concentrations, did not alter the resting membrane potential, whereas higher concentrations evoked perceptible hyperpolarization of the membrane potential (Figs. 1, E–G, and 2). Hyperpolarizing responses to MRS 2365 were unaffected when it was applied in the presence of TTX concentrations known to block neuronal excitability and axonal conduction in the ENS (62). Application of MRS 2365 with the competitive and the selective P2Y₁ receptor antagonist MRS 2179 or MRS 2500 in the tissue bath suppressed the hyperpolarizing responses evoked by MRS 2365 (Figs. 1, C and D, and 2). Responses to 10 μM–1 mM β-NAD were unaffected by the presence of MRS 2500, MRS 2179, or TTX (Figs. 1H and 2).

Fig. 1.

Effects of purinergic P2Y₁ receptor agonists and antagonists and β-NAD on membrane potential of guinea pig colonic circular muscle fibers. A: application of the selective P2Y₁ receptor agonist MRS 2365 evoked hyperpolarization of membrane potential. E: 10 μM β-NAD had no effect for the same preparation. B: tetrodotoxin (TTX) did not suppress hyperpolarizing action of MRS 2365, suggestive of a direct action at the musculature for the same preparation. F: 100 μM β-NAD had no effect. C: hyperpolarizing action of MRS 2365 was suppressed by the P2Y₁ receptor antagonist MRS 2179. G: minimal action of 1 mM β-NAD. D: hyperpolarizing action of MRS 2365 was suppressed by the P2Y₁ receptor antagonist MRS 2500. H: TTX suppressed a small hyperpolarization evoked by a high concentration (1 mM) of β-NAD. Downward deflections are electrotonic potentials evoked by repetitive injection of hyperpolarizing rectangular current pulses. Increased amplitude of the electrotonic potentials indicates increased input resistance, which is an approximation because of the electrical syncytial properties of the musculature.

Fig. 2.

Quantitative data for effects of selective P2Y₁ receptor agonists and antagonists, ATP, and β-NAD on membrane potential in circular muscle fibers of guinea pig colon and human jejunum. A: guinea pig data obtained from impalements in 27 preparations from the same number of animals. B: human data obtained from impalements in 22 preparations from the same number of preparations. *P < 0.05 for suppression of the action of the P2Y₁ receptor agonist MRS 2365 by the selective P2Y₁ receptor antagonist MRS 23179.

Bath application of ATP or MRS 2365 evoked concentration-dependent hyperpolarization of the membrane potential during impalements in 19 circular muscle fibers in 12 preparations from human jejunum (Fig. 3). β-NAD, in micromolar concentrations, did not alter the resting membrane potential, whereas higher (millimolar) concentrations evoked small hyperpolarizing changes in membrane potential (Fig. 2). Potency for the putative P2Y₁ receptor agonists was ranked in the following order: MRS 2365 > ATP >> β-NAD (Fig. 3).

Fig. 3.

Concentration-response relations for effects of β-NAD, the P2Y₁ receptor agonist MRS 2365, and ATP on membrane potential of muscle fibers in human jejunal circular muscle.

As was the case for guinea pig colon, hyperpolarizing responses to MRS 2365 in human jejunum were suppressed by the P2Y₁ receptor antagonists MRS 2179 and MRS 2500 (Fig. 2). Responses to β-NAD were unaffected by the presence of MRS 2500 in 19 preparations obtained from 12 gastric bypass patients.

Smooth muscle IJPs.

Focal electrical stimulation applied to ganglia or interganglionic fiber tracts in the myenteric plexus evoked IJPs in the circular muscle coat of preparations obtained from the small intestine and colon of 25 guinea pigs (18 males and 7 females) and the jejunum of 14 patients (4 males and 10 females) undergoing Roux-en-Y gastric bypass surgery. The IJPs were characteristic biphasic hyperpolarizing potentials comprising an initial fast phase followed by a slowly activating, longer-lasting slow phase reminiscent of IJPs in our earlier reports and those of others (Fig. 4) (23, 34, 65).

Fig. 4.

Inhibitory junction potentials (IJPs) evoked in the circular muscle coat by focal electrical stimulation of intraganglionic fiber tracts in the myenteric division of the enteric nervous system in human jejunal preparations. A₁: control IJP consisting of a rapidly activating component followed by a slowly activating component. A₂: suppression of the IJP by the P2Y₁ receptor agonist MRS 2365 due to hyperpolarization of the membrane potential by MRS 2365 and movement of the membrane potential closer to the equilibrium potential for the IJP. A₃: suppression of the IJP evoked at the resting membrane potential by β-NAD. B₁: suppression of the IJP by the P2Y₁ receptor antagonist MRS 2179. B₂: suppression of the IJP by the P2Y₁ receptor antagonist MRS 2500. B₃: suppression of the IJP by a high concentration of β-NAD. C₁: suppression of the IJP by the high concentration of β-NAD was reversed by preincubation with the adenosine A₁ receptor antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX). C₂: suppression of the IJP by adenosine. C₃: final washout. Preparations were washed between application of each agent or combination of agents.

In view of widespread recognition of purinergic nucleotides as ligands for inhibitory receptors on presynaptic terminals and at neuroeffector junctions (10, 39, 46), we asked the following question: How might β-NAD influence neurotransmission for IJPs in intestinal smooth muscle?

Results were acquired for actions of β-NAD and related purinergic receptor ligands on electrically evoked IJPs in the circular muscle coat of 21 guinea pig colonic preparations (6 proximal, 9 mid, and 7 distal) and 14 human jejunal preparations. We found that ATP, ADP, β-NAD, and adenosine, as well as MRS 2179, MRS 2500, and the adenosine A₁ receptor agonist CCPA, each suppressed the fast phase of IJPs in guinea pig and human preparations (Figs. 4 and 5). Their potency was ranked in the following order: MRS 2500 > MRS 2179 > ADP > ATP > adenosine > CCPA > β-NAD. Potency of ADPR equaled that of β-NAD (Fig. 5). Placement of DPCPX, a potent and selective antagonist for the adenosine A₁ receptor, into the tissue bath prior to application of β-NAD obstructed the action of β-NAD, applied in millimolar concentrations, to suppress electrically evoked IJPs (Fig. 4).

Fig. 5.

Quantitative data for suppression of IJPs by ligands acting at purinergic receptors in guinea pig colon and human jejunum. ADPR, ADP-ribose; CCPA, 2-chloro-N⁶-cyclopentyladenosine.

Intestinal muscle strips.

We prepared strips of intestinal wall from longitudinal and circular orientations for recording of contractile behavior. The mucosa and submucosa were removed from each strip, resulting in circular or longitudinally oriented strips consisting of longitudinal muscle, circular muscle, myenteric plexus, and serosa. The strips were placed in chambers containing Krebs solution and connected to force transducers for purinergic pharmacological analysis of contractile behavior and to electrodes for electrical field stimulation. Twenty-four strips from the colon of 12 guinea pigs, as well as 12 circular and 12 longitudinal strips from 6 jejunal segments from 2 men and 4 women, were studied.

Guinea pig colonic muscle preparations.

Bath application of β-NAD, in the micromolar range, essentially abolished spontaneously occurring contractions, relaxed baseline contractile tension, and suppressed the force of twitch contractions evoked by electrical field stimulation for each of 12 circular strips, from as many animals, relative to controls (Fig. 6B, see Fig. 8, C and D). Application of MRS 2365 or ATP mimicked the action of β-NAD to suppress the force of spontaneously occurring contractions and relax baseline contractile tension (Fig. 6C; see Fig. 8C). Neither MRS 2365 nor ATP suppressed the force of twitchlike contractions evoked by electrical field stimulation (Fig. 6C, see Fig. 8D). Preincubation of the preparations with the P2Y₁ receptor antagonist MRS 2500 blocked the inhibitory action of MRS 2365 on spontaneously occurring contractile activity and basal contractile tension (Fig. 6D). The actions of β-NAD to suppress spontaneously occurring contractions and force of electrically evoked twitch contractions were not reduced by blockade of P2Y₁ receptors by MRS 2500 (Fig. 6, B and E; see Fig. 8, C and D), whereas MRS 2500 did block the action of MRS 2365 to suppress spontaneous contractile activity and relax baseline tension (Fig. 6, C and D, see Fig. 8, C and D).

Fig. 6.

Effects of β-NAD on spontaneously occurring contractile activity and twitch contractions of the circular muscle coat evoked by electrical field stimulation (Elect stim) in guinea pig colon. A: spontaneous contractile activity and twitch response to electrical field stimulation in the absence of pharmacological agents (Control). B: exposure to β-NAD relaxed baseline contractile tension, suppressed spontaneous contractile activity, and suppressed electrically evoked twitch response. C: the P2Y₁ receptor agonist MRS 2365 suppressed spontaneous contractile activity and baseline contractile tension, but not the electrically evoked twitch response. D: action of the P2Y₁ receptor agonist MRS 2365 was suppressed by the P2Y₁ receptor antagonist MRS 2500. E: action of β-NAD was unchanged by the presence of the P2Y₁ receptor antagonist MRS 2500. F: presynaptic inhibition of neurotransmitter release by endomorphin-2 (Endomor) suppressed spontaneous contractile activity, relaxed baseline contractile tension, and suppressed the electrically evoked twitch response. G: blockade of adenosine A₁ receptors by DPCPX had no effect on the action of the P2Y₁ receptor agonist MRS 2365. H: blockade of adenosine A₁ receptors by DPCPX prevented the action of β-NAD to suppress spontaneous contractile activity, relax baseline contractile tension, and suppress the electrically evoked twitch response. I: neural blockade by TTX mimicked the action of β-NAD. J: action of the P2Y₁ receptor agonist MRS 2365 to suppress spontaneously occurring contractile activity was unaffected by TTX. K: the absence of action of micromolar β-NAD when applied in the presence of TTX. L: negligible suppression of baseline contractile tension evoked by millimolar concentration of β-NAD applied in the presence of TTX.

Fig. 8.

Quantitative data for effects of β-NAD on spontaneously occurring contractile activity and twitch contractions of the circular muscle coat evoked by electrical field stimulation in human jejunum and guinea pig colon. A: spontaneous contractile activity in guinea pig colon. β-NAD and the adenosine A₁ receptor agonist CCPA augmented the force of spontaneously occurring contractions. DPCPX, an adenosine A₁ receptor antagonist, suppressed the action of β-NAD. MRS 2500, a P2Y₁ receptor antagonist, did not suppress the action of β-NAD. Suppression by MRS 2365 reflects stimulation of inhibitory P2Y₁ receptors expressed by the musculature. B: electrically evoked twitch contractions in guinea pig colon. β-NAD and the adenosine A₁ receptor agonist CCPA suppressed the force of electrically evoked twitch contractions. DPCPX, an adenosine A₁ receptor antagonist, suppressed the action of β-NAD. MRS 2500, a P2Y₁ receptor antagonist, did not suppress the action of β-NAD. C: spontaneous contractile activity in human jejunum. β-NAD and the adenosine A₁ receptor agonist CCPA reduced the force of spontaneously occurring contractions to levels below baseline tension. DPCPX, an adenosine A₁ receptor antagonist, reversed the action of β-NAD. MRS 2500, a P2Y₁ receptor antagonist, did not suppress the action of β-NAD. Suppression by MRS 2365 reflects stimulation of inhibitory P2Y₁ receptors expressed by the musculature. D: electrically evoked twitch contractions in human jejunum. β-NAD and the adenosine A₁ receptor agonist CCPA suppressed the force of electrically evoked twitch contractions. DPCPX, an adenosine A₁ receptor antagonist, suppressed the action of β-NAD. MRS 2500, a P2Y₁ receptor antagonist, did not suppress the action of β-NAD. *P < 0.05, **P < 0.01 vs. control.

TTX-induced neural blockade suppressed the force of spontaneously occurring contractions, relaxed baseline contractile tension, and blocked electrically evoked contractions in the colon (Fig. 6I). The action of micromolar or millimolar concentrations of β-NAD to suppress the amplitude of spontaneously occurring contractions and to relax basal tension was reduced by TTX (Fig. 6, B and K). Suppression of the amplitude of spontaneously occurring contractions and relaxation of baseline tension, evoked by stimulation of P2Y₁ receptors with MRS 2365, were unaffected by TTX-induced neural blockade (Fig. 6J).

Stimulation of adenosine A₁ receptors with CCPA suppressed the force of spontaneously occurring contractions and the force of twitch contractions evoked by electrical field stimulation in like manner to the action of β-NAD (Fig. 8, C and D). Blockade of adenosine A₁ receptors with DPCPX suppressed the inhibitory action of β-NAD on the force of spontaneously occurring contractions and electrically evoked contractions in the guinea pig colon (Fig. 6, B and H, see Fig. 8, C and D). Blockade of adenosine A₁ receptors with DPCPX had no effect on stimulation of P2Y₁ receptors by MRS 2365 (Fig. 6, C and G).

Endomorphin-2, an endogenous high-affinity μ-opioid receptor agonist, acts presynaptically to suppress neurotransmitter release in the brain, spinal cord, and ENS (52, 66, 67). In view of the presynaptic inhibitory action of adenosine A₁ receptor agonists, we used endomorphin-2 as a comparative agent known to suppress neurotransmission in a manner similar to adenosine. Application of endomorphin-2 mimicked the action of β-NAD to suppress the amplitude of spontaneously occurring contractions, relax basal contractile tension, and suppress the force of twitch contractions evoked by electrical field stimulation (Fig. 6F).

Human jejunal muscle preparations.

Bath application of β-NAD enhanced the force and frequency of spontaneously occurring contractions and suppressed contractile force of twitchlike contractions evoked by electrical field stimulation in 11 of 12 circular strips (Figs. 7, A and B, and 8, A and B). When applied to longitudinal strips, β-NAD had no effect on spontaneously occurring contractions or twitch contractions evoked by electrical field stimulation (data not shown). Application of MRS 2365, ADP, or ATP suppressed the force of spontaneously occurring contractions and reduced baseline tension for the same set of human muscle strips (Figs. 7C and 8A). Neural blockade by TTX enhanced spontaneous contractile activity in a typical manner but did not offset P2Y₁ receptor agonist action of MRS 2365 to suppress spontaneous contractile activity (Fig. 7, H and I) (3, 54). ADP or ATP suppressed the force of twitch contractions evoked by electrical field stimulation by 14.5%, whereas MRS 2365 had no effect on force of twitch contractions evoked by electrical stimulation (Figs. 7C and 8B). The presence of the P2Y₁ receptor antagonist MRS 2500 in the tissue bath did not offset the action of β-NAD to enhance the force of spontaneously occurring contractions (Figs. 7D and 8A). MRS 2500 always prevented the inhibitory action of MRS 2365, ADP, or ATP on spontaneous contractile activity (Figs. 7, C and E, and 8A); quantitative data for ADP and ATP are not presented. β-NAD, in micromolar concentrations, suppressed the force of twitchlike contractile responses to electrical field stimulation, and this action was not reversed by the P2Y₁ receptor antagonist MRS 2500 (Figs. 7, A, B, and D, and 8B).

Fig. 7.

Effects of β-NAD on spontaneously occurring contractile activity and twitch contractions of the circular muscle coat evoked by electrical field stimulation in human jejunum. A: spontaneous contractile activity and twitch response to electrical field stimulation in the absence of pharmacological agents (Control). B: exposure to β-NAD augmented spontaneous contractile activity and suppressed electrically evoked twitch response. C: the P2Y₁ receptor agonist MRS 2365 suppressed spontaneous contractile activity, but not the electrically evoked twitch response. D: action of β-NAD was unchanged by the presence of the P2Y₁ receptor antagonist MRS 2500. E: action of the P2Y₁ receptor agonist MRS 2365 was suppressed by the P2Y₁ receptor antagonist MRS 2500. F: ADP suppressed spontaneous contractile activity, but not the electrically evoked twitch response. G: exposure to the adenosine A₁ receptor agonist CCPA enhanced spontaneously occurring contractile activity and suppressed the electrically evoked twitch response. H: neural blockade by TTX enhanced spontaneously occurring contractile activity and suppressed the electrically evoked twitch response. I: action of the P2Y₁ receptor agonist MRS 2365 to suppress spontaneously occurring contractile activity was unaffected by TTX. J: absence of action of β-NAD applied in the presence of TTX.

The actions of β-NAD to enhance the force of spontaneously occurring contractions and to suppress the force of circular muscle contractions evoked by electrical field stimulation were mimicked by bath application of CCPA, a specific agonist for the adenosine A₁ receptor (Figs. 7G and 8B) (35). Preincubation with DPCPX, a potent and selective antagonist at the adenosine A₁ receptor, reversed the action of β-NAD to enhance the force of spontaneously occurring contractions and to suppress the force of circular muscle contractions evoked by electrical field stimulation (Fig. 7, A and B) (40, 41).

DISCUSSION

Smooth muscle electrophysiology.

We found that junction potentials in the circular muscle coat, evoked by focal electrical stimulation of musculomotor neurons in the myenteric plexus, consisted of IJPs and excitatory junction potentials and were basically similar to those reported in several studies by others in which they were evoked by transmural electrical field stimulation (1, 16, 34, 49, 51, 65). IJPs in the present study consisted of two phases: an initial larger-amplitude, rapidly activating hyperpolarizing component (fast IJP) followed by a smaller and longer-lasting hyperpolarizing component (slow IJP). Slow IJPs were sometimes overridden by the larger fast IJP and were uncovered by blockade of the fast IJPs with the P2Y₁ receptor antagonists MRS 2179 and MRS 2500. Actions of ATP and the P2Y₁ receptor agonist MRS 2365 included hyperpolarization of the membrane potential, which in general was the same as reported by others (18, 22, 30, 33, 42, 56). Failure of TTX to suppress the hyperpolarizing responses to MRS 2365 was consistent with a direct action of the P2Y₁ receptor agonist at receptors expressed in the musculature, rather than excitation of inhibitory musculomotor neurons, known also to express excitatory P2Y₁ receptors (20, 26, 31). Data for suppression of MRS 2365-evoked membrane hyperpolarization by the selective P2Y₁ receptor antagonists MRS 2179 and MRS 2500 in guinea pig colon and human jejunum add to existing evidence for involvement of P2Y₁ receptors in enteric-purinergic inhibitory neuromuscular transmission (20, 26, 31).

In view of debate on roles for ATP and β-NAD as inhibitory neuromuscular transmitters in the GI tract, we compared their actions on the membrane potential of muscle cells in the circular coat of guinea pig colon and human jejunum (23, 29, 44, 50). We found that β-NAD was considerably less potent than ATP or MRS 2365 as a hyperpolarizing agonist in the guinea pig and human preparations. Concentrations in the millimolar range were required to evoke hyperpolarizing responses by β-NAD. The IC₅₀ for β-NAD was ∼12 times greater than that for ATP and >100 times greater than that for MRS 2365 (Fig. 3). Whereas the P2Y₁ receptor antagonists suppressed the hyperpolarizing action of MRS 2365, neither MRS 2179 nor MRS 2500 suppressed the hyperpolarizing action of β-NAD. Overall, our data for the guinea pig colon and human jejunum are consistent with data reported by Gallego et al. (22) for human colon and are inconsistent with a role for β-NAD as an inhibitory neurotransmitter released by the motor innervation of the circular muscle coat.

As expected, the P2Y₁ receptor antagonists MRS 2179 and MRS 2500 suppressed fast IJPs in guinea pig and human preparations in our study. Suppression in these cases was most likely due to blockade of P2Y₁ receptors. The amplitude of fast IJPs was slightly reduced in the presence of very high concentrations of β-NAD. The reduction of amplitude of the IJPs in the presence of high concentrations of β-NAD might have reflected β-NAD-induced hyperpolarization and movement of the membrane potential toward the equilibrium potential for the IJP and/or a prejunctional action to suppress transmitter release. ADPR, which, like β-NAD, accumulates in the bathing medium in a TTX-sensitive manner during electrical stimulation, also slightly suppressed the amplitude of IJPs when applied in relatively high concentrations (32). Suppression of IJPs by high concentrations of β-NAD did not occur in the presence of the potent and selective adenosine A₁ receptor antagonist DPCPX.

Along with β-NAD and ADPR, accumulation of adenosine has been measured in superfusate from intestinal preparations during electrical stimulation of intramural nerves (32). This occurs because ATP, after release into the extracellular milieu, is hydrolyzed rapidly to ADP, AMP, and adenosine by a group of membrane-bound nucleotidases known as ecto-ATPases (48, 68). Suppression of the IJPs by high concentrations of β-NAD did not occur when DPCPX, a potent and selective adenosine A₁ receptor antagonist, was present in the bathing medium. Adenosine itself and the selective adenosine A₁ receptor agonist CCPA always suppressed or abolished stimulus-evoked IJPs in guinea pig and human preparations. Strong evidence that inhibitory adenosine A₁ receptors are expressed on most presynaptic axon terminals in the ENS and on motor nerve terminals innervating GI effectors (9, 10, 15, 20, 21, 67) raises the following question: Might β-NAD act to suppress release of the neurotransmitters that evoke IJPs?

A possible explanation for mimicry of adenosine A₁ receptor agonists by β-NAD is that β-NAD acts at inhibitory prejunctional adenosine A₁ receptors to suppress action potential-evoked release of the inhibitory neurotransmitter at the junction. Adenosine A₁ receptors are known to be expressed by presynaptic axonal terminals in the ENS, and stimulation of these receptors is known to suppress neurotransmitter release from the terminal (10, 13). A prejunctional inhibitory action of β-NAD is a likely explanation for its action to suppress neurotransmission at intestinal neuromuscular junctions.

Pharmacological actions of methylxanthines, such as DPCPX, can involve inhibition of intracellular phosphodiesterase activities, as well as act as a receptor ligand (55). Action of this nature for β-NAD in the present study is unlikely, because suppression of phosphodiesterase activity in ENS musculomotor neurons elevates, rather than suppresses, excitability (47).

Muscle preparations.

We used spontaneously occurring contractile activity and responses to electrical field stimulation as “readouts” for evaluation of purine neuropharmacology in relation to β-NAD. Spontaneously occurring contractile amplitude, baseline tension, and contractile frequency are parameters that can be measured as reflections of levels of excitatory and inhibitory ENS motor neuronal activity. Firing frequencies of excitatory and inhibitory musculomotor neurons in concert determine the minute-to-minute amplitude of spontaneous contractions and, thereby, provide a basis for evaluation of pharmacological agents. For example, when ongoing firing of inhibitory motor neurons is the predominant neural activity, any agent (e.g., TTX) that blocks firing of the neurons removes inhibition from the autogenic smooth muscle and increases in the amplitude of spontaneous contractions appear on the record (Fig. 7H) (54, 60, 61).

Electrical field stimulation was expected to fire most of the neurons in the preparations and release their neurotransmitters. Parameters that were used (pulse duration, current, and stimulus frequency) restricted neuronal activation to firing of the neuronal cell bodies, because the fine (∼0.5-μm) nonmyelinated axons of enteric motor neurons were not fired by the stimulus parameters used because of the high chronaxie of the nerve fibers (7). Chronaxies in smooth muscles in tissue or organ baths require large currents to reach action potential threshold. Our stimulators were unable to provide sufficient current to depolarize the muscle itself to threshold. Therefore, all contractile responses to electrical field stimulation reflected firing of excitatory musculomotor neurons, as confirmed by TTX blockade.

Action of β-NAD to blunt twitchlike contractile responses to electrical field stimulation was similar for guinea pig colon and human jejunum. Amplitude of spontaneously occurring contractions and baseline tension were reduced by exposure to β-NAD in guinea pig colon. Human jejunum differed, in that amplitude of spontaneous contractions and baseline tension were increased by exposure to β-NAD. Reduction, by β-NAD, of the amplitude of electrically evoked twitch contractions suggests that β-NAD acted to suppress release of excitatory neurotransmitters at the neuromuscular junctions. In view of its action on stimulus-evoked contractions, it appears likely that elevation of the amplitude of spontaneously occurring contractions in human jejunum reflected suppression of ongoing release of inhibitory neurotransmitter at neuromuscular junctions in a manner similar to the excitatory action resulting from blockade of ongoing activity of inhibitory motor neurons by TTX. The reduction in baseline tension and amplitude of spontaneously occurring contractions by β-NAD in guinea pig colon can be explained by suppression of release of excitatory neurotransmitters at the neuromuscular junctions, if it is assumed that the excitatory musculomotor neurons were firing spontaneously prior to application of β-NAD. None of these actions of β-NAD can be attributed to stimulation of P2Y₁ receptors by β-NAD, because they were unaffected by the P2Y₁ receptor antagonists MRS 2500 and MRS 2179.

Stimulation of adenosine A₁ receptors by the established adenosine A₁ receptor agonist CCPA mirrored the actions of β-NAD in guinea pig and human preparations. Blockade of adenosine A₁ receptors by DPCPX offset the action of β-NAD to reduce the force of twitch contractions evoked by electrical field stimulation. Adenosine A₁ receptor blockade also reversed effects of β-NAD on spontaneously occurring contractile behavior in the guinea pig. Effects of β-NAD on spontaneous contractile behavior took place at the level of ENS neurons, because they did not occur in the presence of neural blockade with TTX. Effects of β-NAD in human preparations differed, in that its action to elevate the amplitude of spontaneously occurring contractile activity mimicked the effect of neural blockade with TTX. Subpopulations of inhibitory musculomotor neurons fire continuously in intestinal preparations, such as those in the present study, and, thereby, exert continuous suppression of contractile activity of the autogenic musculature (59–61). The action of β-NAD, which appears to be release of the musculature from ongoing inhibition, and its action to suppress contractions evoked by electrical field stimulation could result from suppression of excitability of inhibitory musculomotor neurons or suppression of neurotransmitter release at enteric synapses and/or neuromuscular junctions. In this respect, β-NAD appears to suppress neurotransmitter release in a manner similar to stimulation of the A₁ receptor subtype by adenosine.

Endomorphin-2, another physiological inhibitor of neurotransmitter release at axonal terminals, was evaluated for comparison with observed actions of β-NAD (52, 66, 67). Endomorphin-2 is known to suppress contractions evoked by electrical field stimulation in rat ileum (52). β-NAD acted in an identical manner for stimulus-evoked contractions in human jejunum and guinea pig colon. Moreover, behavior of endomorphin-2 resembles suppression of the release of neurotransmitter from inhibitory musculomotor neurons during the descending phase and the release of neurotransmitter from excitatory musculomotor neurons during the ascending phase of the multisynaptic propulsive motor reflex in the rat intestine (52). Suppression of the release of inhibitory neurotransmitter by endomorphin-2 during the descending phase of peristaltic propulsion is suggestive of the action of β-NAD to mimic the excitatory effects of TTX on spontaneously occurring contractile activity in intestinal preparations.

Conclusion.

Our results are consistent with a hypothesis that the rapidly activating phase of the IJPs in the small and large intestinal circular muscles is mediated by the purinergic P2Y₁ receptor subclass. The results do not support an inhibitory neurotransmitter role for β-NAD at intestinal neuromuscular junctions. The data suggest that β-NAD is a ligand for the adenosine A₁ receptor subtype expressed by neurons in the ENS. Action of β-NAD on intestinal motility emerges from suppression of neurotransmitter release at synapses in the ENS microcircuitry, at ENS neuromuscular junctions, and by suppression of excitability of ENS ganglion cells, each of which is mediated by stimulation of the adenosine A₁ receptor subtype.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants RO1 DK-37238 and KO8 DK-060468.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

G.-D.W., S.L., Y.X., B.J.N., D.J.M., and J.D.W. are responsible for conception and design of the research; G.-D.W., X.-Y.W., S.L., F.Z., M. Q, and J.D.W. performed the experiments; G.-D.W., X.-Y.W., S.L., Y.X., B.J.N., D.J.M., and J.D.W. analyzed the data; G.-D.W., S.L., Y.X., B.J.N., D.J.M., and J.D.W. interpreted the results of the experiments; G.-D.W., X.-Y.W., S.L., Y.X., B.J.N., D.J.M., F.Z., M.Q., and J.D.W. approved the final version of the manuscript; S.L. and J.D.W. edited and revised the manuscript; J.D.W. prepared the figures; J.D.W. drafted the manuscript.