SEARCH

SEARCH BY CITATION

Keywords:

  • inhibitory junctivvon potential;
  • MRS2500;
  • P2Y1;
  • P2Y1 receptors;
  • β-NAD.

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

Background  Pharmacological studies using selective P2Y1 antagonists, such as MRS2500, and studies with P2Y1−/− knockout mice have demonstrated that purinergic neuromuscular transmission is mediated by P2Y1 receptors in the colon. The aim of the present study was to test whether P2Y1 receptors are involved in purinergic neurotransmission in the antrum and cecum.

Methods  Microelectrode recordings were performed on strips from the antrum and cecum of wild type animals (WT) and P2Y1−/− mice.

Key Results  In the antrum, no differences in resting membrane potential and slow wave activity were observed between groups. In WT animals, electrical field stimulation elicited a MRS2500-sensitive inhibitory junction potential (IJP). In P2Y1−/− mice, a nitrergic IJP (Nω-nitro-l-arginine-sensitive), but not a purinergic IJP was recorded. This IJP was equivalent to the response obtained in strips from WT animals previously incubated with MRS2500. Similar results were obtained in the cecum: 1- the purinergic IJP (MRS2500-sensitive) recorded in WT animals was absent in P2Y1−/− mice 2- nitrergic neurotransmission was preserved in both groups. Moreover, 1- spontaneous IJP (MRS2500-sensitive) could be recorded in WT, but not in P2Y1−/− mice 2- MRS2365 a P2Y1 agonist caused smooth muscle hyperpolarization in WT, but not in P2Y1−/− animals, and 3- β-NAD caused smooth muscle hyperpolarization both in WT and P2Y1−/− animals.

Conclusions & Inferences  1- P2Y1 receptor is the general mechanism of purinergic inhibition in the gastrointestinal tract, 2- P2Y1−/− mouse is a useful animal model to study selective impairment of purinergic neurotransmission and 3- P2Y1−/− mouse might help in the identification of purinergic neurotransmitter(s).


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

In the gastrointestinal (GI) tract, inhibitory neuromuscular transmission is mainly mediated by NO1,2 and ATP or a related purine.3 The release of inhibitory neurotransmitters elicits an inhibitory junction potential (IJP) that usually displays two consecutive components i.e. a fast purinergic component followed by a slow nitrergic component.4,5 This co-transmission is present in most of the regions of the GI tract and in the vast majority of the species including human colon,6 rat colon,7 rat internal anal sphincter 8, and guinea pig ileum.9 Exceptions are the human esophagus and lower esophageal sphincter, where the IJP is mainly nitrergic.10

The characterization of the co-transmission process involves two possible approaches: (i) the use of pharmacological tools using agonist/antagonists for a specific receptor or selective inhibitors of the pathways involved in the synthesis of the mediator and (ii) genetically modified animals that lack a specific protein necessary for the neurotransmission of a specific pathway. Therefore, the nitrergic neuromuscular transmission is blocked by Nω-nitro-l-arginine (l-NNA) and ODQ11,12 and absent both in knockout mice for nNOS 13–15 and in knockout mice for guanylate cyclase (GC).16 All these results demonstrate that nitric oxide causes postjunctional responses through cGMP. Regarding the purinergic pathway, MRS2179, a P2Y1 antagonist, abolishes the IJP at concentrations that differ between the human colon6 and murine colon,7 suggesting that other receptor subtypes might also contribute to the purinergic response.17,18 Recent studies with more potent antagonists of the P2Y1 receptor, such as MRS2279 or MRS2500, have confirmed the crucial role for this receptor in human colonic neurotransmission.19 This pharmacological approach has been recently validated using colon strips obtained from P2Y1 knockout mice, where purinergic neurotransmission is absent.20,21 This experimental approach has been considered a ‘resolution and concordance in dissecting the compound IJP’.22 The nature of the purinergic neurotransmitter is nowadays under deep debate.23–26 This is due to the fact that, in addition to ATP or related purine such as ADP,3 other purines as β-NAD23,24 might be bioactive and could contribute to the purinergic IJP. It is important to notice that the majority of these studies have been performed exclusively in the colon, an organ specialized in storage and elimination of waste. It is unknown if in other areas of the GI tract, with different functions, the same mechanisms occur. Accordingly, the aim of the present study was to investigate the inhibitory neurotransmission in the gastric antrum and cecum of P2Y1 knockout mice and to compare data with putative agonists and antagonists of the P2Y1 receptor. In the present article, we confirmed the crucial involvement of P2Y1 receptor as a general mechanism of purinergic inhibition probably in the whole GI tract.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

Animals and tissue samples

Male C57BL/6J wild type (WT; n = 14) and P2Y1-deficient (B6.129P2-P2ry1tm1Bhk/J) mice (P2Y1−/−; n = 5) 10- to 15-week-old were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Male CD1 animals (n = 5) (Charles River Laboratories, L’Arbresle Cedex, France) were also used to perform specific experiments. Animals were housed under controlled conditions: temperature 22 ± 2 °C, humidity 55 ± 10%, 12 : 12-h light–dark cycle and access to water and food ad libitum. Animals were killed by cervical dislocation. Both the stomach and cecum were quickly removed and placed in carbogenated (95% O2 and 5% CO2), ice cold physiological saline solution. Muscle strips devoid of mucosa and submucosa were prepared (8 × 2 mm) from the gastric antrum and cecum. All procedures were approved by the Ethics Committee of the Universitat Autònoma de Barcelona.

Muscle bath studies

Muscle strips, circularly oriented, were mounted in a 10 mL organ bath containing carbogenated physiological saline solution maintained at 37 ± 1 °C. Motility was measured using an isometric force transducer (Harvard VF-1 Harvard Apparatus Inc., Holliston, MA, USA) connected to a computer through an amplifier. Data were digitalized (25 Hz) using Data 2001 software (Panlab, Barcelona, Spain) coupled to an A/D converter installed in the computer. A tension of 0.5 g was applied and tissues were allowed to equilibrate for 1 h. After this period, carbachol 10 μmol L−1 was added and strips displayed phasic activity. The release of inhibitory neurotransmitters was studied by using electrical field stimulation (EFS) applied for 1 min; pulse duration 0.5 ms; frequency 3 Hz; amplitude 40 V. The area under the curve (AUC) of contractions from the baseline was measured to estimate the mechanical activity and the result was expressed in grams per minute (g min−1) and normalized to calculate the percentage of inhibition during EFS.

Intracellular microelectrode recordings

The tissue was pinned with the circular muscle layer facing upwards in a Sylgard-coated chamber and continuously perfused with carbogenated physiological saline solution at 37 ± 1 °C and was allowed to equilibrate for 1 h. Circular smooth muscle cells were impaled with glass microelectrodes filled with 3 mol L−1 KCl (30–60 MΩ of tip resistance). Membrane potential was measured by using standard electrometer Duo773 (WPI Inc., Sarasota, FL, USA). Tracings were displayed on an oscilloscope 4026 (Racal-Dana Ltd., Windsor, UK) and simultaneously digitalized (100 Hz) with PowerLab 4/30 system and Chart 5 software for Windows (both from ADInstruments, Castle Hill, NSW, Australia). To stabilize impalements, experiments were performed in the presence of nifedipine (1 μmol L−1). Slow waves activity was recorded in gastric antrum samples. Frequency, amplitude, and duration of slow waves were measured and expressed in cycles per minute (c.p.m.), millivolts (mV), and seconds (s), respectively. In these samples, resting membrane potential (RMP) was established as the average of the membrane potential between each slow wave (in mV). In the cecum, the spontaneous inhibitory neural tone was characterized as previously described in the colon.27 Briefly, the resting membrane potential (expressed in mV) was estimated as the most probable bin of the frequency distribution of the membrane potential (0.1 mV bins; 30–60 s recordings). Spontaneous inhibitory junction potentials were evaluated by calculating the mean standard deviation (SD) of the distribution of the membrane potential: SD of the recording inside the cell minus SD of the recording outside the cell (expressed in mV). Inhibitory junction potential were also elicited by EFS using the following parameters: 1 Gastric antrum: single pulses, pulse duration 0.6 ms and increasing voltage (8, 12, 16, 20, 24, 28, 32, 36, and 40 V);2 Cecum: protocol 1 – single pulses, pulse duration 0.3 ms and increasing voltage (8, 12, 16, 20, 24, 28, 32, 36, and 40 V); protocol 2 – train stimulation during 0.5 s performed at 20 Hz, 0.3 ms pulse duration and increasing voltage (8, 12, 16, 20, 24, 28, 32, 36, and 40 V); protocol 3 and 4 – train stimuli of 5 s duration (supramaximal amplitude and pulse duration 0.3 ms) performed at 1 Hz, 5 pulses (protocol 3) and 5 Hz, 25 pulses (protocol 4). Inhibitory junction potential amplitude was measured from the RMP and expressed in mV. Train pulses at 1 Hz elicited five consecutive hyperpolarizations which were defined by their amplitude. For train stimuli at 5 Hz, the fast component of the IJP was assessed by measuring the maximum amplitude of the IJP during the initial pulses and the amplitude of the slow component was assessed at 2.5 and 3.75 s after the beginning of the stimulus.

Solutions and drugs

The composition of the physiological saline solution was (in mmol L−1) glucose 10.10; NaCl 115.48; NaHCO3 21.90; KCl 4.61; NaH2PO4 1.14; CaCl2 2.50, and MgSO4 1.16 (pH 7.3–7.4). In all the experiments, phentolamine, atropine, and propranolol (1 μmol L−1) were added to the physiological saline solution to block α- and β-adrenoreceptors and muscarinic receptors. The following drugs were used: atropine sulfate, β-nicotinamide adenine dinucleotide sodium salt (β-NAD), nifedipine, l-NNA, -nitro-l-arginine methyl ester (l-NAME), carbamoylcholine chloride (carbachol), phentolamine, sodium nitroprusside (SNP) (Sigma Chemicals, St. Louis, MO, USA); 2′-deoxy-N6-methyladenosine 3′,5′-bisphosphate tetrasodium salt (MRS2179), [[(1R,2R,3S,4R,5S)-4-[6-Amino-2-(methylthio)-9H-purin-9-yl]-2,3 dihydroxybicyclo[3.1.0]hex-1-yl]methyl] diphosphoric acid mono ester trisodium salt (MRS2365), (1R,2S,4S,5S)-4-[2-Iodo-6-(methylamino)-9H-purin-9-yl]-2-(phosphonooxy)bicyclo[3.1.0]hexane-1-methanol dihydrogen phosphate ester diammonium salt (MRS2500), propanolol (Tocris, Bristol, UK). Stock solutions were made by dissolving drugs in distilled water except for nifedipine, which was dissolved in 96% ethanol, and l-NNA, which was dissolved in physiological saline solution by sonication.

Data analysis and statistics

Two-way analysis of variance (anova) followed by Bonferroni’s multiple comparison test was used to compare differences on IJP between WT and P2Y1−/− mice and between drug treatments. To compare the effect of MRS2179 and MRS2500 on the IJP, IC50 was calculated using conventional sigmoid concentration-response curve with variable slope. Statistical analysis was performed using a two-way anova followed by Bonferroni’s multiple comparison test. Paired Student’s t-test was used to evaluate: (i) differences in the resting membrane potential before and after drug addition; (ii) the effect of MRS2500 on the hyperpolarization induced by MRS2365, β-NAD and SNP. One-way anova was used to evaluate the drug effect on spontaneous IJP.

Data are expressed as mean ± SEM. A P < 0.05 was considered statistically significant. ‘n’ values indicate the number of samples from different animals. Statistical analysis and curve fit were performed with GraphPad Prism version 4.00, (GraphPad Software, San Diego, CA, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

Nitrergic and P2Y1 mediated purinergic relaxation in the gastric antrum and cecum

Muscle bath experiments performed in CD1 animals (n = 5) revealed that carbachol (10 μmol L−1) induced an increase in motility in circularly oriented strips (Fig. 1). Carbachol induced irregular contractions in the cecum whereas in the antrum regular contractions at a frequency of (4.8 ± 0.5 min−1) were recorded. These results are consistent with the slow wave activity recorded in the antrum and the lack of slow waves recorded in the cecum (see below). Electrical field stimulation-induced relaxation was completely antagonized by the sequential addition of l-NNA (1 mmol L−1) and MRS2500 (1 μmol L−1) showing that both NO and a purine acting on P2Y1 receptors are responsible for smooth muscle relaxation (Fig. 1). Interestingly, in both tissues, the fast IJP was l-NNA (1 mmol L−1) insensitive and MRS2500 (1 μmol L−1) sensitive (not shown) confirming the P2Y1 nature of the receptor. For a proper comparison (identical genetic background), the rest of the following electrophysiological experiments were performed in male C57BL/6J (WT) and B6.129P2-P2ry1tm1Bhk/J) P2Y1−/− mice.

image

Figure 1.  (A) Muscle bath recordings showing the spontaneous motility induced by carbachol (10 μmol L−1) in strips circularly oriented from the mouse cecum (top tracing) and antrum (bottom tracing). Electrical field stimulation (EFS) caused an inhibition of this spontaneous motility. The sequential incubation with Nω-nitro-l-arginine (1 mmol L−1) and MRS2500 (1 μmol L−1) was necessary to block the relaxation and (B) Histograms showing the percentage of area under the curve inhibition during the EFS before and after drug addition (anovaP < 0.005 in the cecum and P < 0.0001 in the antrum).

Download figure to PowerPoint

Gastric antrum

Slow waves and inhibitory tone in WT and P2Y1−/− mice  Circular smooth muscle cells had a mean RMP measured at the bottom of slow waves of −53.1 ± 1.0 mV (n = 8) in WT mice and −51.4 ± 1.8 mV (n = 5; ns) in P2Y1−/− mice. No differences between WT and P2Y1−/− animals were measured in terms of frequency (WT: 6.5 ± 0.5 vs P2Y1−/−: 6.4 ± 0.2 c.p.m.; ns), duration (WT: 7.8 ± 0.3 vs P2Y1−/−: 7.4 ± 0.5 s; ns), and amplitude (WT: 9.2 ± 1.3 vs P2Y1−/−: 9.2 ± 1.4 mV; ns; Figs 2 and 3) of slow waves. Slow wave activity was not modified by l-NNA (1 mmol L−1) or MRS2500 (1 μmol L−1), both in WT and P2Y1−/− animals.

image

Figure 2.  Representative recordings of inhibitory junction potentials induced by four protocols of electrical field stimulation (EFS) in the gastric antrum (A) and cecum (B) of WT (left) and P2Y1−/− (right). From top to bottom: ‘Protocol 1’: Single pulses at increasing voltage of EFS (8, 12, 16, 20, 24, 28, 32, 36, and 40 V) in the gastric antrum and cecum ‘Protocol 2’: Train of 20 Hz and 0.5 s increasing voltage of EFS (8, 12, 16, 20, 24, 28, 32, 36, and 40 V) ‘Protocol 3’: Train stimuli of 5 s duration and 1 Hz (5 pulses) at supramaximal voltage of stimulation and ‘Protocol 4’: Train stimuli of 5 s duration and 5 Hz (25 pulses) at supramaximal voltage of stimulation in the cecum. Plot graphs and histograms showing a comparison between WT and P2Y1−/− mice. All values are mean ± SEM (n = 8 WT and 5 P2Y1−/− in the antrum and n = 11 WT and 5 P2Y1−/− in the cecum). Comparison between the inhibitory junction potential amplitude from WT and P2Y1−/− mice was significant (Two-way anova: P < 0.0001).

Download figure to PowerPoint

image

Figure 3.  Representative recordings of a WT mouse (left and right) and P2Y1−/− mouse gastric antrum (center). Effect of the sequential addition of Nω-nitro-l-arginine (l-NNA) and MRS2500 and MRS2500 and l-NNA on the inhibitory junction potential (IJP) (single pulse of electrical field stimulation) of WT animals and effect of l-NNA on the IJP of P2Y1−/− (Notice that no IJP was recorded after incubation with l-NNA in P2Y1−/− mice). Bottom: Graphs showing the amplitude of the response at increasing voltages of stimulation in WT animals (n = 8) and P2Y1−/− mice (n = 5). All values are mean ± SEM. Comparison between the IJP amplitude from WT and P2Y1−/− mice was significant (Two-way anova: P < 0.0001).

Download figure to PowerPoint

Inhibitory junction potential induced by EFS in WT and P2Y1−/− mice  In WT animals, single pulse stimulation caused an IJP with an average amplitude of −10.5 ± 1.0 mV at supramaximal voltage (n = 8). In contrast, in P2Y1−/− mice, IJP were comparatively smaller in amplitude (supramaximal voltage: −1.8 ± 0.6 mV; n = 5; P < 0.001) (Fig. 3 top). Nω-nitro-l-arginine (1 mmol L−1) did not modify the IJP in WT mice (−9.0 ± 1.2 mV supramaximal voltage; n = 4; ns). Further addition of MRS2500 (1 μmol L−1) completely abolished the IJP (Fig. 3 left). In contrast, in P2Y1−/− mice, the IJP was completely abolished by l-NNA (Fig. 3 center). In a different subset of experiments performed in WT animals, MRS2500 (1 μmol L−1) partially reduced IJP which showed similar features to which observed in P2Y1−/− mice (−2.0 ± 0.8 mV at supramaximal voltage; n = 4; P < 0.001). Further addition of l-NNA (1 mmol L−1) abolished IJP (Fig. 3 right).

Effect of MRS2365, ß-NAD, and SNP on smooth muscle RMP in WT and P2Y1−/− mice  Responses to exogenous purinergic and nitrergic agonists were tested on WT and P2Y1−/− mice. MRS2365 (1 μmol L−1), a selective P2Y1 agonist, only caused a slight smooth muscle hyperpolarization in WT animals (Control: −57.4 ± 2.3 vs MRS2365: −59.6 ± 2.3; n = 7; ns) and had no effect in P2Y1−/− mice. β-NAD 1 mmol L−1 did not modify RMP in WT mice (Control: −55.5 ± 2.1 vsβ-NAD: −57.2 ± 2.1; n = 6; ns). However, it clearly reduced slow waves. Finally, the nitric oxide donor SNP (10 μmol L−1) slightly hyperpolarized (Control: −50.6 ± 1.1 vs SNP: −55.4 ± 1.2; n = 4; P < 0.05) smooth muscle cells and completely abolished slow wave activity. Similar results were obtained in P2Y1−/− mice for β-NAD and SNP (data not shown).

Cecum

Inhibitory neural tone in WT and P2Y1−/− mice  Electrophysiological recordings from cecum did not display slow wave activity even in the absence of nifedipine. To characterize the neurotransmission process, nifedpine was used to stabilize the impalements. In WT mice, circular smooth muscle cells had a mean RMP of −42.3 ± 1.5 mV (n = 13); no differences were observed when compared to P2Y1−/− mice (−40.4 ± 2.0 mV; n = 5, ns). Both in WT and P2Y1−/− mice, l-NNA caused a slight smooth muscle depolarization (WT: 2.9 ± 0.8 mV, n = 6; P2Y1−/−: 3.9 ± 0.8 mV, n = 4; P < 0.05). This depolarization was also confirmed in preliminary experiments with l-NAME (1 mmol L−1) in CD1 animals excluding the possibility of non-specific effects of l-NNA in RMP (data not shown). Subsequent addition of MRS2500 1 μmol L−1 in WT animals (n = 6) did not cause any further change in RMP. In WT mice, ‘spontaneous’ IJP were often recorded increasing the ‘noise’ of the recording. The SD of the data was 0.5 ± 0.1 mV (n = 13). In WT animals, spontaneous IJP were still present after l-NNA incubation and were reduced after the incubation with MRS2500 1 μmol L−1 (from 0.5 ± 0.1 to 0.1 ± 0.1 mV SD; n = 6, P < 0.01).

Inhibitory junction potential induced by EFS in WT and P2Y1−/− mice  Four different protocols of EFS were used to assess the electrophysiological responses: single pulse trains increasing the voltage of stimulation (from 8 to 40 V), short trains of 0.5 s and 20 Hz increasing the voltage of stimulation (from 8 to 40 V), trains of 5 s at 1 Hz (supramaximal voltage), and trains of 5 s at 5 Hz (supramaximal voltage). Responses are shown in Fig. 2. In WT animals, EFS with a single pulse and short trains of 20 Hz elicited a fast IJP reaching an amplitude of 13.1 ± 0.9 mV for the single pulse (1.3 ± 0.1 s of duration) (n = 11); and an amplitude of 24.7 ± 1.3 mV for the short train (20 Hz; 0.5 s) (1.4 ± 0.1 s of duration) (n = 11). In contrast, in P2Y1−/− mice single pulses elicited a small IJP reaching an amplitude of 1.9 ± 0.9 mV (0.9 ± 0.4 s of duration) (n = 5); and the short train elicited a long lasting IJP of 6.23 ± 2 mV and 6.08 ± 2.4 s of duration (n = 5) (Fig. 2). In WT animals, trains of 5 s at 1 Hz elicited five consecutive fast IJP (P1: 12.3 ± 1.1 mV; P2: 10.6 ± 1.0 mV; P3: 13.0 ± 1.0 mV; P4: 13.2 ± 0.9 mV; P5: 14.3 ± 1.2 mV) (n = 11). In contrast, the response in P2Y1−/− mice consisted of five small IJP that progressively increased in amplitude (P1: 1.1 ± 0.5 mV; P2: 1.2 ± 0.5 mV; P3: 2.3 ± 0.8 mV; P4: 2.3 ± 0.7 mV; P5: 3.3 ± 1.0 mV) (Fig. 2).

Trains of 5 s at 5 Hz elicited a fast hyperpolarization in WT animals (IJPf: 16.69 ± 0.99 mV) followed by a sustained hyperpolarization with similar amplitude to the IJPf (IJPs2.5: 16.0 ± 1.1 mV; IJPs3.75: 15.3 ± 0.9 mV) (n = 11) whereas in P2Y1−/− mice, only the slow component could be recorded (IJPf: 0mV; IJPs2.5: 11.2 ± 1.9 mV; IJPs3.75: 11.9 ± 1.7 mV). Comparisons between WT vs P2Y1−/− animals were different in all cases (Two-way anovaP < 0.0001) (Fig. 2).

Nω-nitro-l-arginine (1 mmol L−1) and MRS2500 (1 μmol L−1) were used to investigate the co-transmission process. In WT animals, l-NNA did not modify the amplitude of the fast IJP [single pulse, short train (20 Hz; 0.5 s), 1 and 5 Hz]. Subsequent addition of MRS2500 1 μmol L−1 completely abolished all the electrophysiological responses. In P2Y1−/− mice, l-NNA completely abolished all the electrophysiological responses showing that the MRS2500 sensitive component was absent in these animals (n = 6) (Figs 4 and 5).

image

Figure 4.  Representative recordings of inhibitory junction potentials (IJPs) induced by four protocols of electrical field stimulation in the cecum of WT mice (top) and P2Y1−/− mice (bottom). From left to right ‘Protocol 1”: Single pulses “Protocol 2”: Train of 20 Hz and 0.5 s “Protocol 3”: Train stimuli of 5 s duration and 1 Hz (5 pulses) and “Protocol 4”: Train stimuli of 5 s duration and 5 Hz (25 pulses) all of them at supramaximal voltage of stimulation. “Protocol A”. Sequential addition of Nω-nitro-l-arginine (l-NNA) (1 mmol L−1) and MRS2500 (1 μmol L−1) and Protocol B. Sequential addition of MRS2500 (1 μmol L−1) and l-NNA (1 mmol L−1). Notice that no IJP was recorded after incubation with l-NNA in tissue from P2Y1−/− mice.

Download figure to PowerPoint

image

Figure 5.  (A) Graphs of plots and histograms from WT mice showing the effect of sequential addition of Nω-nitro-l-arginine (l-NNA) (1 mmol L−1) and MRS2500 (1 μmol L−1) on the inhibitory junction potential (IJP) amplitude. (B) Graphs of plots and histograms from P2Y1−/− animals showing the effect of l-NNA (1 mmol L−1) on the IJP. (C) Graphs of plots and histograms from WT mice showing the effect of sequential addition of MRS2500 (1 μmol L−1) and l-NNA (1 mmol L−1). From top to bottom “Protocol 1”: Single pulses “Protocol 2”: Train of 20 Hz and 0.5 s both of them increasing the voltage of stimulation “Protocol 3”: Train stimuli of 5 s duration and 1 Hz (5 pulses), and “Protocol 4”: Train stimuli of 5 s duration and 5 Hz (25 pulses) both of them at supramaximal voltage of stimulation.

Download figure to PowerPoint

In a different subset of experiments in WT animals, MRS2500 (1 μmol L−1) was added prior to l-NNA to evaluate the nitrergic component and to compare WT with P2Y1−/− mice. In WT animals, MRS2500 (1 μmol L−1) clearly reduced the amplitude of IJP elicited by the different stimulation protocols (n = 4; P < 0.0001 in all the EFS protocols) (Figs 4 and 5). Interestingly, the nitrergic component of the IJP was significantly lower in WT animals than in the P2Y1−/− mice in two of the four different protocols of EFS: in the short train (20 Hz; 0.5 s) (2.8 ± 0.6 mV n = 4 WT vs 6.2 ± 2.1 mV n = 5 P2Y1−/− Two-way anovaP < 0.001) and the train of 5 Hz during 5 s (IJPs 2.5 = 6.5 ± 2.0 mV, IJP3.75 = 6.6 ± 2.3 mV n = 4 WT vs IJPs 2.5 = 11.2 ± 1.9 mV, IJP3.75 = 11.9 ± 1.7 mV n = 5 P2Y1−/− Two-way anovaP < 0.05). Further addition of l-NNA (1 mmol L−1) abolished all the electrophysiological responses (Figs 4 and 5).

Concentration–response curves of MRS2500 and MRS2179 on the IJPf  Concentration–response curves were performed in the presence of l-NNA to evaluate the effect of MRS2179 and MRS2500 on the supramaximal IJPf (using EFS protocol 2, the short train of 20 Hz and 0.5 s). The IC50 for MRS2500 was 20.1 nmol L−1 (LogIC50 from −7.9 to −7.5 with 95% CI) and the IC50 for MRS2179 was 8.8 μmol L−1 (LogIC50 from −5.1 to −5.0 with 95% CI). It was necessary for a concentration of 20 μmol L−1 of MRS2179 to completely block the IJPf whereas only a small proportion of the IJPf remained at the concentration of MRS2500 of 0.1 μmol L−1 (Fig. 6).

image

Figure 6.  Representative tracings showing the effect of MRS2179 (A) and of MRS2500 (B) on the inhibitory junction potential elicited by electrical field stimulation (trains of 20 Hz, 0.5 s and supramaximal voltage) in the mice cecum. (C) Concentration–response curves of MRS2500 and MRS2179. Tissue was previously incubated with Nω-nitro-l-arginine (1 mmol L−1).

Download figure to PowerPoint

Effect of MRS2365, ß-NAD, and SNP on smooth muscle RMP in WT and P2Y1−/− mice  Responses to exogenous purinergic and nitrergic agonists were tested on WT and P2Y1−/− mice. The nitric oxide donor SNP (10 μmol L−1) induced a smooth muscle hyperpolarization both in WT (12.9 ± 4.1; n = 4) and P2Y1−/− animals (11.0 ± 0.2; n = 2).The selective P2Y1 agonist MRS2365 1 μmol L−1 induced a smooth muscle hyperpolarization (−17.0 ± 1.8 mV; n = 8) and an increase in the SD of the recording (from 0.4 ± 0.1 to 0.9 ± 0.2 SD; n = 8, P < 0.01). Both effects were almost completely blocked by the P2Y1 antagonist MRS2500 1 μmol L−1 (−1.4 ± 0.9 mV; n = 8 P < 0.0001 and 0.2 ± 0.1 SD; n = 8, P < 0.01). In contrast, in P2Y1−/− animals, MRS2365 1 μmol L−1 (n = 5) did not induce any change in smooth muscle RMP or an increase in the SD (Fig. 7). β-NAD 1 mmol L−1 hyperpolarized (−17.6 ± 2.5 mV, n = 7) smooth muscle cells of WT animals. The hyperpolarization was slightly reduced by preincubation with MRS2500 although often a residual effect was observed (−11.62 ± 1.70 mV, n = 7; ns). In P2Y1−/− animals, β-NAD induced a smooth muscle hyperpolarization (−20.09 ± 3.14 mV; n = 5; ns from WT) (Fig. 7). β-NAD 1 mmol L−1 did not induce an increase in the SD (both in WT and P2Y1−/− animals). Posterior experiments using CD1 mice were performed to evaluate the effect of lower concentrations of β-NAD. β-NAD concentration dependently hyperpolarized the tissue (10 μmol L−1: 1.6 ± 0.5 mV and 100 μmol L−1: 7.8 ± 1.0 mV both n = 4), but did not modify SD.

image

Figure 7.  (A) Representative recordings from the cecum showing the hyperpolarization induced by MRS2365 (1 μmol L−1) in a WT mouse (left) which was blocked by MRS2500 (1 μmol L−1) and was absent in P2Y1−/− animals (right). (B) hyperpolarization induced by β-NAD (1 mmol L−1) in WT animals (left), the effect of β-NAD (1 mmol L−1) in the presence of MRS2500 (1 μmol L−1) and in P2Y1−/− animals (right) (C) hyperpolarization induced by SNP (10 μmol L−1) in WT and P2Y1−/−.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

Genetically modified mice, lacking pathways/receptors, are useful biological tools that when combined with an appropriate pharmacological approach allows to characterize neuromuscular transmission in the GI tract. In this study, we confirmed in the gastric antrum and in the cecum previous observations made in the human and rat colon6,7,19 using P2Y1 antagonists (MRS2179, MRS2279, and MRS2500) that completely blocked purinergic neuromuscular transmission. Previous data both in the rat colon7 and human colon19 suggested that rank of order of potencies was MRS2500 > MRS2279 > MRS2179. This pharmacological approach has recently been validated with P2Y1 knockout mice (P2Y1−/−) where colonic purinergic neuromuscular transmission is absent.20,21 In this study, we demonstrate the absence of purinergic IJP in the antrum and cecum of P2Y1−/− mice. We propose that P2Y1 receptor activation is a general mechanism for nerve mediated purinergic inhibitory responses in the whole GI tract.

In the mice stomach, the fundic region has a stable membrane potential whereas cyclic slow waves were recorded in the antrum (6–7 c.p.m.). This slow wave activity generates cyclic contractions (4–6 c.p.m.) when the tissue is stimulated with carbachol. Cessation of cyclic contractions induced by EFS was reversed by sequential addition of l-NNA and MRS2500 as previously reported in the colon.6,7,20 This pharmacological approach suggests that a co-transmission between ATP/NO is essential to explain gastric relaxation. It is important to notice that animals that lack ICC do not generate proper slow wave activity.28 The resting membrane potential of smooth muscle cells in the antrum was more hyperpolarized compared with other areas of the GI tract including the fundus.29 Neither recordings from P2Y1−/− nor from strips incubated with MRS2500 had any change in slow wave activity or RMP, showing that P2Y1 receptors are unlikely involved in the regulation of RMP and the onset or development of slow waves. The fact that RMP is quite hyperpolarized in this tissue might explain the little effect observed after incubation with the P2Y1 agonist (MRS2365) or the NO donor (SNP). The low gradient between RMP and the equilibrium for potassium ions might explain why only a small smooth muscle hyperpolarization was recorded. This is also consistent with the smaller IJP measured in the gastric antrum reaching only 8–10 mV in amplitude. IJP were reduced by tissue incubation with MRS2500 in WT animals, but not in P2Y1−/− mice. In both cases, the residual non-purinergic response was abolished by l-NNA, and therefore was nitrergic. All these results support the existence of an ATP/NO co-transmission. A crucial role for nNOS has been previously demonstrated in the fundic region in nNOS−/− mice13 and the present work demonstrates that the purinergic response is absent in the antrum of P2Y1−/− mice.

In a previous work, it was suggested that P2Y1 receptors act as prejunctional modulators of ATP release in the mouse cecum, having a minor contribution at the postjunctional level.18 These findings differ from this study. The discrepancies between both studies could be attributed to the relative potency of the different P2Y1 receptors antagonists. MRS2500 is more potent than MRS2179 both in human, rat colon7,19 and mouse cecum (present study). These results explain the lack of effect of MRS2179 when low concentrations of the antagonist are used.18 The fact that P2Y1−/− mice are lacking purinergic neurotransmission rules out possible non-specific effects for MRS2500 and confirms a major role for P2Y1 receptors in purinergic neuromuscular transmission in the cecum.

The inhibitory effect of MRS2500 was tested using several types of EFS [single pulses, short trains, and long trains at two frequencies (1 and 5Hz)], far from being redundant, these EFS protocols, were necessary to demonstrate that, despite the absence of purinergic neurotransmission, the nitrergic innervation was intact. In fact, in some of the stimuli we observed an increase in nitrergic responses in P2Y1−/− mice, suggesting a possible ‘compensatory’ mechanism. However, 1- SNP-induced hyperpolarization was exactly the same in WT and P2Y1−/− mice; 2- RMP did not differ between WT and P2Y1−/− mice, and 3- l-NNA equally depolarized smooth muscle cells in WT and P2Y1−/−mice. All these results suggest that if a ‘compensatory’ mechanism occurs, it should be at the prejunctional level and can be detected only after EFS. However, a complete compensation is not possible because functions differ between neurotransmitters. In the human colon, a more transient and phasic relaxation had been attributed to purines, whereas NO could be responsible for a more prolonged or tonic relaxation.30 In the cecum, as previously shown in the colon, an ongoing release of NO (but not purines) regulates the RMP. i.e. Nitric Oxide synthesis inhibitors depolarizes smooth muscle cells both in WT and P2Y1−/−mice; and spontaneous IJPs are blocked by MRS2500 and reduced in P2Y1−/− suggesting that purines are ‘spontaneously’ released from inhibitory motor neurons and are responsible for the spontaneous IJPs.12,20,27

Recent studies suggest that β-NAD mimics better than ATP/ADP, the endogenous neurotransmitter.23,24 In the cecum, both SNP and MRS2365 caused a smooth muscle hyperpolarization of about 13 and 17 mV, respectively, therefore we infer that this is the appropriated tissue to test the putative involvement of β-NAD as a neurotransmitter. MRS2365, the selective P2Y1 agonist, induced: (i) a smooth muscle hyperpolarization that was blocked by MRS2500 in WT animals; (ii) an increase in ‘spontaneous’ IJP in WT animals; and (iii) both effects (hyperpolarization and increase in spontaneous IJP) were absent in P2Y1−/− animals. These results show that selective exogenous P2Y1 agonist properly mimics the endogenous response. In contrast, β-NAD caused a smooth muscle hyperpolarization, but this response was only slightly reduced by MRS2500. β-NAD did not increase spontaneous IJP and β-NAD-induced hyperpolarization was still recorded in P2Y1−/− mice. Our experimental approach cannot conclude about the nature of the endogenous neurotransmitter, but this study reveals non-related P2Y1 effects of β-NAD as previously described in the guinea pig taenia coli31 and colon.19 All together, our results suggest that β-NAD does not fulfill the criteria to be considered and endogenous purinergic neurotransmitter. Further studies are needed to evaluate the effect of other purines such as ATP/ADP or β-NAD metabolites.

Genetically modified animals can also help in the study of the location of receptors that participate in inhibitory neurotransmission. Interestingly, in W/Wv mice antrum, lacking ICC-IM, the nitrergic neurotransmission is absent32, but in the colon nitrergic neurotransmission is only partially impaired in Ws/Ws rats.33 The fact that nitrergic innervation is still present in animals that lack specific smooth muscle GC demonstrate a possible role for ICC-IM in nitrergic neurotransmission34 although the contrary (absence of CG in ICC-IM) has to be confirmed. Fibroblast-like cells express PDGRFα receptor, P2Y1 receptors, and SK3 channels.35–37 These cells are interesting candidates to transduce purinergic signals from neurons to smooth muscle. However, Fibroblast-like cells also express GC.35 Whether PDGRFα + cells in coordination with ICC can participate in the co-transmission process (NO and ATP) and how the stimulus can be translated to the smooth muscle cells remains unknown. Selective deletions in specific cells might be helpful in future studies to clarify the mechanism.34

We conclude that the purinergic neuromuscular transmission is mediated by P2Y1 in the antrum, cecum, and colon.20,21 P2Y1 receptors are a general mechanism mediating purinergic inhibition throughout the GI tract. Moreover, P2Y1−/− animals might be interesting models to investigate possible diseases where the purinergic mechanism is impaired.38

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

The authors would like to thank Antonio Acosta and Claudia Arenas for their technical assistance.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

This work has been funded by the following grant: BFU2009-11118. Víctor Gil and Míriam Martínez-Cutillas are supported by the Ministerio de Ciencia e Innovación (Spain) (AP2007-01583) and (AP2010-2224), respectively. Diana Gallego is supported by the Instituto de Salud Carlos III, Centro de Investigación Biomédica en red de enfermedades hepáticas y digestivas (CIBERehd).

Author Contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References

VG designed the experiments, performed the research, analyzed the data, and contributed to writing of the manuscript; MMC and NM performed some experiments and analyzed some of the data; MTM contributed to writing of the manuscript; MJ designed the experiments and contributed to writing of the manuscript; DG designed the experiments, performed the research, analyzed the data, and wrote the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Funding
  9. Disclosure
  10. Author Contributions
  11. References