Dopamine induces inhibitory effects on the circular muscle contractility of mouse distal colon via D1- and D2-like receptors

Dopamine (DA) acts as gut motility modulator, via D1- and D2-like receptors, but its effective role is far from being clear. Since alterations of the dopaminergic system could lead to gastrointestinal dysfunctions, a characterization of the enteric dopaminergic system is mandatory. In this study, we investigated the role of DA and D1- and D2-like receptors in the contractility of the circular muscle of mouse distal colon by organ-bath technique. DA caused relaxation in carbachol-precontracted circular muscle strips, sensitive to domperidone, D2-like receptor antagonist, and mimicked by bromocriptine, D2-like receptor agonist. 7-Chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride (SCH-23390), D1-like receptor antagonist, neural toxins, L-NAME (nitric oxide (NO) synthase inhibitor), 2′-deoxy-N6-methyl adenosine 3′,5′-diphosphate diammonium salt (MRS 2179), purinergic P2Y1 antagonist, or adrenergic antagonists were ineffective. DA also reduced the amplitude of neurally evoked cholinergic contractions. The effect was mimicked by (±)-1-phenyl-2,3,4,5-tetrahydro-(1H)-3-benzazepine-7,8-diol hydrobromide (SKF-38393), D1-like receptor agonist and antagonized by SCH-23390, MRS 2179, or L-NAME. Western blotting analysis determined the expression of DA receptor proteins in mouse distal colon. Notably, SCH-23390 per se induced an increase in amplitude of spontaneous and neurally evoked cholinergic contractions, unaffected by neural blockers, L-NAME, MRS 2179, muscarinic, adrenergic, or D2-like receptor antagonists. Indeed, SCH-23390-induced effects were antagonized by an adenylyl cyclase blocker. In conclusion, DA inhibits colonic motility in mice via D2- and D1-like receptors, the latter reducing acetylcholine release from enteric neurons, involving nitrergic and purinergic systems. Whether constitutively active D1-like receptors, linked to adenylyl cyclase pathway, are involved in a tonic inhibitory control of colonic contractility is questioned.


Introduction
The exact definition of the role of dopaminergic signaling in the gastrointestinal (GI) motility is still a major challenge. Enteric dopaminergic neurons, expressing both the dopamine (DA)-synthesizing enzyme tyrosine hydroxylase and dopamine transporter but not the norepinephrine -synthesizing enzyme, have been reported within the enteric nervous system (ENS) in both animal models and humans [2,15]. Effects of DA depend on the activation of five receptor subtypes, constituting two families: D1-like family, including D1 and D5 receptors, and D2-like family, including D2, D3, and D4 receptors [9]. All DA receptor subtypes have been found within the GI tract [16,26], being D1-like receptors mainly located at postjunctional level and D2-like receptors located both at preand postjunctional level. DA primarily induces inhibitory effects on GI motility from the esophagus to the colon [25,30]. DA causes relaxation in guinea-pig and rat jejunum [19,20] and reduction of the spontaneous contractions in mouse small Michelangelo Auteri and Maria Grazia Zizzo equally contributed to this work intestine via activation of D1-like receptors [33,35]. In murine colon, a combination of D1-and D2-like receptor antagonists increases the muscle tone, the amplitude of spontaneous contractions, and the neurally evoked contractions, supporting a role of endogenous DA as a tonic negative modulator of colonic mechanical activity [28]. However, subsequent studies failed to state the specific contribution of D1-and D2-like receptors. D2 knockout (KO) mice display increased colonic propulsive activity and decreased total GI transit [16]; the D2 receptor antagonist itopride accelerates peristalsis in guineapig colon [17] and the D2 receptor agonist bromocriptine reduces intestinal transit in mice [10], strongly suggesting an important contribution of D2 receptors in the DA-induced inhibitory effect on GI motility. Indeed, other studies highlighted a major role for D1 receptors in mediating DAdependent basal inhibitory control in the mouse ileum [33], as well as the exclusive involvement of D1 receptors in the DAinduced reduction of phasic contraction in the rat distal colon [31].
Currently, the prokinetic effects of antidopaminergic drugs (i.e. domperidone, clebopride) are widely exploited for the treatment of different motor disorders of the upper GI tract, including dyspepsia and gastroparesis [18,24]. However, the increased GI transit and colonic motility in D2 KO mice [16] as well as the effects of DA and dopaminergic drugs on the motility of the lower digestive tract highlighted that the pharmacological modulation of enteric dopaminergic signaling could be of importance throughout the entire GI tract and not only in the proximal gut.
Therefore, since the function of DA in gut motility is far from being clear and that large intestine dysmotility characterized different GI motor disorders, the aim of this study was to investigate, in mice, the effects induced by DA on spontaneous and neurally evoked contractions of colonic circular muscle preparations, particularly exploring the possible specific role of D1-and D2-like receptors.

Animals
Experiments were performed using adult male mice (C57BL/ 6) weighing about 25 g obtained from Charles River Laboratories (Calco-Lecco, Italy). Adult C57BL/6 mice have been demonstrated to be a good model for studying intestinal motility [31,32]. Animals were housed in a specific pathogenfree environment and kept under environmentally controlled condition (ambient temperature 24°C, humidity 40% and 12 h light/dark cycle) and housed at four to a cage with food and water ad libitum. Procedures involving animals and their care were conducted in conformity the Italian D.L. no. 116 of 27 January 1992 and subsequent variations, and the European directives (2010/63/EU). All applicable international, national, and/or institutional guidelines for the care and use of animals were followed [11]. No other methods to perform the described experiments (3Rs) were found.

Functional studies in vitro
Mice were randomly selected for experiments and euthanized using isoflurane anesthesia followed by cervical dislocation. After laparotomy, the colon was rapidly excised, placed in Krebs solution (mM: NaCl 119; KCl 4.5; MgSO 4 2.5; NaHCO 3 25; KH 2 PO 4 1.2, CaCl 2 2.5, glucose 11.1). Then, a segment of distal colon (about 5 mm proximal to the anus) of approximately 3.5 cm length was obtained, and the contents of the excised segments were gently flushed out with Krebs solution.
The contractile activity of distal colonic circular smooth muscle was recorded as previously described by Auteri et al. [3]. Briefly, segments of distal colon were opened along the mesenteric border and pinned mucosa side up. To avoid the influence of the substances released from the mucosa, it was removed by sharp dissection under a microscope, and fullthickness circular muscle strips (10 mm in length) were suspended in a four-channel organ bath containing 10 ml of oxygenated (95% O 2 and 5% CO 2 ) Krebs solution maintained to 37°C. The distal end of each strip was tied to an organ holder, and the proximal end was secured with a silk thread to an isometric force transducer (FORT 25, Ugo Basile, Biological Research Apparatus, Comerio VA, Italy). Mechanical activity was amplified and digitized via an analogue/digital interface (Quad Bridge and PowerLab/400, AD Instruments, Ugo Basile, Biological Research Apparatus, Comerio VA, Italy), prior being acquired onto a personal computer. The preparations were subjected to an initial tension about 5 mN and were allowed to equilibrate for at least 30 min.
After the equilibration time, preparations were challenged either with 10 μM carbachol (CCh) or with 1 μM isoproterenol (Iso) for 2 min until stable responses were obtained. Since preliminary experiments showed that DA induced only a weak inhibitory effect on the mechanical activity of circular colonic strips at basal tone (data not shown), preparations were precontracted with 10 μΜ CCh. The effects of such a concentration of CCh were blocked by atropine (1 μM) and not affected by the Na + voltage gated channel blocker, TTX (1 μM), indicating a direct action of the agonist on muscarinic receptors located at post-junctional level.
Concentration-response curves for dopamine or SKF-38393 and bromocriptine, D1-like and D2 like receptor agonist, respectively, were constructed on CCh-precontracted strips by non-cumulative addition of the drugs, applied for approximately 3 min at 20 min intervals, in the absence or in the presence of 1 μM SCH-23390 or 5 μM domperidone, D1-like and D2-like receptor antagonists, respectively. The dopaminergic antagonists were allowed to maintain contact with the tissue for at least 30 min before eventually repeating the curve to the dopaminergic agonists. Time control experiments showed that a second curve to the agonists was reproducible. Each preparation was tested with a single antagonist, except when otherwise stated. In order to evaluate the related mechanism of action, a submaximal dose of dopamine (100 μM) was tested on precontracted strips in the presence of different inhibitors, left in contact to the tissue for at least 20 min. In a separate set of experiments, due to the effects of the D1 receptor antagonist on the spontaneous activity, SCH-23390 (3 μM) was tested after pretreatment of 30 min with the neural blockers, TTX or ω-conotoxin GVIA, ω-agatoxin TK and SNX-482,blocker of N-P/Q-and R-type calcium channels, or atropine, muscarinic receptor antagonist, or L-NAME, NO synthase inhibitor, or MRS 2179, purinergic P2Y1 receptor antagonist, prazosin, yohimbine, propanolol, and SR 59230A, α1-, α2-, β1/β2, and β3 antagonist, respectively, or 2′,3′dideoxyadenosine (DDA), an adenylyl cyclase inhibitor. Concentrations of the drugs used were determined from our previous experiments [33][34][35].
Lastly, to test the effects of dopamine on the neural evoked responses, electrical field stimulation (EFS) was applied from a Grass S88 electrical stimulator (Grass Instruments Co., Quincy, MA, USA) through a stimulus isolation unit (SIU5) using direct coupling. Stimuli (0.5 ms, 10 V for 10 s) were delivered, via a pair of platinum plate electrodes, in trains at the of 4 Hz frequency, which was chosen to specifically activate cholinergic neurons [3,34]. Dopamine or dopamine receptor agonists were added to the organ bath, and the EFS was repeated after 2 min. Antagonists or blockers tested were left in contact to the tissue for a period of 20-30 min, before testing EFS.

Data and statistical analysis
All the data were analyzed by operators, blinded to the protocol assignment. Data are given as means ± SEM; n in the results section refers to the number of animals on which observations were made. At least five individual experiments were carried out for each independent protocol, and the exact number is reported in the figure legends. The amplitude of the relaxation induced by dopamine or by dopaminergic agonists was reported as a percentage of the effect induced by 1 μM Iso taken as 100%. Responses to the drugs in the absence or presence of the different antagonists were fitted to sigmoid curves (Prism 4.0, GraphPad, San Diego, CA), and EC 50 values with 95% confidence limits (CLs) were determined from these curves. Contractile responses to EFS were expressed as a percentage of the amplitude of contraction induced by 10 μM CCh taken as 100%. Statistical analysis was performed by means of analysis of variance followed by the Bonferroni's post hoc test, when appropriate. A probability value of 0.05 was regarded as significant.

Results
Effects of dopamine on spontaneous colonic circular muscle contractions In CCh-precontracted circular muscle strips, exogenous application of DA (1-300 μM) produced a concentrationdependent relaxant effect, which persisted throughout the contact time and was reversible after washing out (EC 50 = 13.3 μM, 95% CL 11.0-16.1 μM; n = 15). The amplitude of the relaxation in response to 300 μM DA was about 3 mN (Fig. 1a, b).
Lastly, Western Blotting analysis releveled the expression of dopamine receptor proteins in mouse distal colon (Fig. 3).

Analysis of the effect of D1-like receptor antagonist on distal colon contractility
The observation that SCH-23390, D1-like receptor antagonist, per se induced an excitatory effects on the mechanical activity, while activation of D1 receptors had no effect on distal colon contractility, aimed us to further investigate about the specificity of the response. The excitatory effect, consisting in an increase in the amplitude of the spontaneous activity, was concentration-dependent with a threshold at 1 μM and reversible upon washout (Fig. 5a). At the dose of 10 μM, SCH-23390 caused also an increase in the basal tone of about 1.50 mN (Fig. 5a). The neural blockers TTX (1 μM), the blockers of N-P/Q-, or R-type Ca 2+ channels ω-conotoxin  1 μM), ω-agatoxin TK, (0.1 μM) respectively, the muscarinic receptor antagonist, atropine (1 μM), the NO synthase inhibitor, L-NAME (100 μM), or the purinergic P2Y1 receptor antagonist, MRS 2179 (1 μM) were ineffective (Fig. 5b). Moreover, the excitatory effects induced by SCH-23390 persisted in the presence of the α1, α2, β1/β2, and β3 adrenergic receptor antagonists, prazosin (1 μM), yohimbine (1 μM), propranolol (1 μM), and SR 59230A (100 nM) (Fig. 5b). Indeed, the effects of SCH-23390 on the amplitude of spontaneous contractions and on the muscle tone were significantly reduced in the presence of an adenylyl cyclase inhibitor, DDA (10 μM) (Fig. 5a, b). Lastly, the excitatory effects induced by SCH-23390 persisted in the presence of domperidone (Fig. 5b).

Discussion
Results of the present study indicate DA as a negative modulator of colonic motility in mice through stimulation of D1and D2-like receptors. Activation of D2-like receptors induces a net inhibitory effect on colonic circular muscle spontaneous activity, while activation of D1-like receptors decreases acetylcholine (ACh) release within the ENS, via a pathway involving the enteric inhibitory neurotransmitters, namely NO and purines. Accumulating data demonstrated the potential importance of the enteric dopaminergic system in the modulation of GI motility, but the effective role and functional significance of DA and the contribution of its receptors are far from being clear. Previous studies have reported both excitatory and inhibitory effects of DA on colonic motility via activation of D2-like receptors [4,14]. DA is able to stimulate rectosigmoid motility in humans, being such effect antagonized by domperidone, D2-like receptor antagonist [29]. However, most of subsequent studies addressed DA as a major negative modulator of GI motility. DA inhibits electrically induced contractions of colonic longitudinal muscle strips in mice [28] and rats [31], via both D1-and D2-like receptor activation, and colonic peristaltic activity in guinea-pig [17] and rat [14]. Furthermore, DA seems to be also implicated in the opioid-induced reduction of intestinal transit through its interaction with enteric serotoninergic and GABAergic systems [10].
Our results showed that dopamine receptor proteins are expressed in mouse distal colon, and DA itself is able to affect muscular contractility. In detail, in CCh pre-contracted strips, exogenous DA induced a dose-dependent muscular relaxation sensitive to domperidone, D2-like receptor antagonists and mimicked by bromocriptine, D2-like receptor agonist. D1-like receptor antagonist, SCH-23390, did not affect the relaxation induced by DA nor SKF-38393, D1-like receptor agonist, was able to mimic DA inhibitory effect. All together, these results indicate a pivotal role played by D2-like receptors in the relaxation induced by DA, as suggested in other preparations [16,35]. Although we are aware that agonists may relax with different magnitude preparations precontracted with some agents (such as KCl) or with others (such as carbachol), this analysis was behind our scope. Domperidone was unable per se to modify spontaneous mechanical activity excluding a tonic inhibitory action of DA through D2-like receptors. This observation is in contrast to the suggestion of Li et al. [16] which reported that knock-out mice for D2 receptors show an increase in the rate of total gastrointestinal transit and a regional increase in colonic motility. However, results obtained in mutant mice could also depend on loss of central dopaminergic control, reflecting in the observed impairment of gut functions. The inhibitory effect induced by D2-like receptor activation on colonic mechanical activity is not dependent on enteric neuronal action potential or on modulation of neurotransmitter release since it was modified neither by the blocker of Na + voltage-dependent channels nor by the blockers of R, P/Q, and N-type Ca 2+ channels, which have a substantial role in the release of transmitters from enteric neurons [27]. Since the main inhibitory agents mediating relaxation in mouse colon, i.e., NO and purines can be released also by non-neural cells [5,8], we tested the effect induced by DA also in the presence of L-NAME, a NO synthase inhibitor, and MRS 2179, antagonist of purinergic P2Y1 receptors, main purinergic receptors involved in the muscular relaxation. Data obtained indicate that nor NO nor purines acting on P2Y1 receptors are involved in DA-induced relaxation. In addition, although different studies pointed out that some of DA-induced effects could be mediated via the activation of adrenergic receptors [1,7], in our experimental conditions, we can exclude this hypothesis, as DA responses were unaffected by pretreatment with αand β-adrenergic antagonists. Thus, in mouse colon, DA exerted its effects via the activation of its own dopaminergic receptors, namely D2-like receptors.
DA was also able to inhibit the EFS-induced cholinergic contractions, and its effect seems to be mediated by D1-like activation, as already observed in mouse small intestine [33], since the DA inhibitory effect was antagonized by the D1-like receptor antagonist, SCH-23390, and mimicked by SKF-38393, D1-like receptor agonist. DA agents failed to affect the direct muscular cholinergic contraction induced by carbachol, indicating that the sensitivity of the muscle to muscarinic activation was not changed and thus suggesting that the inhibition of neurally evoked cholinergic contractions involves prejunctional mechanisms. ATP and purines, main inhibitory cotransmitters in the circular muscle of mouse colon [21,23], may cause inhibition of intestinal motility acting not only directly on the smooth muscle cells but also indirectly by  1 μM, n = 4), ω-agatoxin TK (0.1 μM, n = 5), with the muscarinic receptor antagonist, atropine (1 μM, n = 5), the NO synthase inhibitor, L-NAME, (100 μM, n = 5), the purinergic P2Y1 receptor antagonist, MRS 2179 (1 μM, n = 5), with the adrenoceptor antagonists, prazosin (1 μM, n = 5), yohimbine (1 μM, n = 5), propanolol (1 μM, n = 5) or SR59230A (100 nM, n = 5), α1, α2, β1/ β2, or β3 antagonists respectively, with the adenylyl cyclase inhibitor, DDA (10 μM, n = 6) or domperidone, and D2 receptor antagonist (5 μM, n = 5). Data are means ± SEM and are expressed in absolute value. The graphed value for the control bar is the mean of the control data obtained before each treatment. *P < 0.05 when compared to the respective own control according to one-way ANOVAs followed by the Bonferroni's post hoc test inhibiting the release of excitatory neurotransmitters such as acetylcholine [6,22]. Dopaminergic inhibitory effects were antagonized by L-NAME, NO synthase inhibitor, and attenuated by the blocker of P2Y1 receptors, MRS 2179, suggesting that, in circular muscle of mouse colon, D1-like receptors could down-regulate ACh release from enteric nerves via activation of purinergic and nitrergic systems. In addition, DA has been reported to modulate the cholinergic transmission in gut preparation via adrenergic receptors located on intramural cholinergic neurons, as observed in rat and in canine gastric fundus [12,13]. However, none of the adrenoceptor antagonists used was able to affect the response induced by DA on EFS, leading us to exclude, once more, an involvement of adrenergic receptor in the DA-induced effects on cholinergic transmission.
Lastly, one intriguing result from our study was the observation that SCH-23390, D1-like receptor antagonist, per se increased basal tone and the amplitude of the spontaneous contractions, being highly indicative for a modulatory role of endogenous DA on gut motility as already reported [28], but activation of D1 receptors by the selective D1-like receptor agonist had no effect on distal mouse colon spontaneous contractility. SCH-23390 is able to antagonize DA and SKF-38393, D1-like receptor agonist, effects on the neurally evoked cholinergic contraction, confirming its ability as D1like receptor antagonist in mouse intestine [3]. The SCH-23390-induced effect was readily reversible and dosedependent in a concentration range normally used in the study on rodent intestine, suggesting a specificity of the response. Moreover, differently from what observed in mouse ileum [33], the excitatory effects induced by SCH-23390 are not antagonized by the neural blockers TTX or ω-conotoxin GVIA, SNX-482, ω-agatoxin TK, blockers of presynaptic Ca 2+ channels used in this study, excluding an interference on receptors located on enteric neurons tonically releasing inhibitory neurotransmitter, as NO and purines, or reducing excitatory neurotransmitter, as ACh. Even the possibility that the increase in the contractility observed in the presence of SCH-23390 can be ascribed to an effect on other receptors, as D2-like receptors or on adrenoreceptors can be discarded since, in our preparation, none of the antagonists for the above mentioned receptors inhibited the SCH-23390-induced excitatory effects. Constitutive activity for D1-like receptors has been established [32]. The sensitivity of the excitatory effects of SCH-23390 to the adenylyl cyclase blocker may support our suspicion, since it is well-known that activation of D1-like receptors initiates a cascade of intracellular events including cyclic adenosine monophosphate pathway [30,35]. However, this hypothesis is weakened by the observation that the selective D1-like receptor agonist SKF-38393, which has been shown to be an efficacious agonist in mouse intestine in this as in other previous studies [30,35], was without any effect on the spontaneous activity. Whether or not D1-like receptors have undergone to conformational changes modifying their pharmacological properties deserve attention, and further experiments are needed.
In conclusion, our work demonstrates that, in mouse colon, exogenous DA induces inhibitory effects on the circular muscle mechanical activity via prejunctional inhibitory D1-and postjunctional D2-like receptors. D1-like receptor activation inhibited cholinergic neural responses through enteric nitrergic and purinergic pathways, in turn dampening ACh release. Remain to determine if constitutively active D1-like receptors, linked to adenylyl cyclase pathways, are involved in a tonic inhibitory control of colonic contractility, under physiological conditions.