- Email: [email protected]
Role of galanin receptor 1 in peristaltic activity in the guinea pig ileum
Neuroscience 125 (2004) 103–112
ROLE OF GALANIN RECEPTOR 1 IN PERISTALTIC ACTIVITY IN THE GUINEA PIG ILEUM C. STERNINI,a,b* L. ANSELMI,a,b,c S. GUERRINI,a,b1 E. CERVIO,c T. PHAM,a,b B. BALESTRA,c R. VICINI,c P. BAIARDI,d G.-L. D’AGOSTINOe AND M. TONINIc
receptors; GAL-R1 mediates the low potency phase. The reduced peristalsis efficiency could be explained by inhibition of the cholinergic drive, whereas the decreased compliance is probably due to inhibition of descending neurons and/or to the activation of an excitatory muscular receptor. Endogenous galanin does not appear to affect neuronal pathways subserving peristalsis in physiologic conditions via GAL-R1. © 2004 IBRO. Published by Elsevier Ltd. All rights reserved.
a CURE Digestive Diseases Research Center, Digestive Diseases Division, Building 115, Room 224, Veterans Administration Greater Los Angeles Healthcare System, 11301 Wilshire Boulevard, Los Angeles, CA 90073, USA b Departments of Medicine and Neurobiology, University of California Los Angeles, David Geffen School of Medicine, Los Angeles, CA 90095, USA
Key words: cholinergic transmission, descending myenteric neurons, ascending myenteric neurons, longitudinal muscle– myenteric plexus preparations, noncholinergic, nonadrenergic inhibitory pathways.
c
Department of Physiological and Pharmacological Sciences, University of Pavia, Pavia, Italy
d
S. Maugeri Foundation, University of Pavia, Pavia, Italy
e
Department of Experimental and Applied Pharmacology, University of Pavia, Pavia, Italy
Galanin is a brain– gut neuropeptide that exerts a variety of cellular functions in the nervous system, including neurotransmitter and hormone release, nociception, spinal reflexes and feeding behavior (Rattan, 1991; Bartfai et al., 1992; Bedecs et al., 1995). In the gastrointestinal tract, galanin is localized to myenteric and submucosal neurons and to fibers projecting to the gut wall (Ekblad et al., 1985b; Melander et al., 1985; Furness et al., 1987), and it affects motility and secretion mainly acting as an inhibitory neuromodulator (Rattan, 1991). The effects of galanin on gastrointestinal smooth muscle are both stimulatory and inhibitory and are the result of a direct myogenic effect and a nerve-mediated effect involving the release of other transmitters (Ekblad et al., 1985a; Fox et al., 1986; Yau et al., 1986; Muramatsu and Yanaihara, 1988; Bauer et al., 1989; Mulholland et al., 1992; Akehira et al., 1995; Gu et al., 1995). For instance, in the guinea-pig intestine, galanin inhibits the electrically induced contraction mediated by endogenous acetylcholine and substance P, and acetylcholine release (Ekblad et al., 1985a; Fox et al., 1986; Yau et al., 1986; Muramatsu and Yanaihara, 1988; Bauer et al., 1989; Mulholland et al., 1992; Akehira et al., 1995). Galanin also hyperpolarizes myenteric neurons and suppresses their excitability (Palmer et al., 1986; Tamura et al., 1987). Together, these pharmacological studies indicate that galanin inhibits the excitatory neuroneuronal and neuromuscular transmission in the guinea-pig ileum, and it exerts direct effects on smooth muscle cells. Galanin effects are mediated by distinct receptors that activate multiple second messenger pathways to affect cell activity (Branchek et al., 1998, 2000; Floren et al., 2000). Three G protein-coupled receptors for galanin have been cloned, named galanin receptor 1 (GAL-R1), GAL-R2 and GAL-R3, which have distinct pharmacological profiles and activate different intracellular signaling pathways (Branchek et al., 2000; Floren et al., 2000). GAL-R1 inhibits forskolin-stimulated cAMP production by coupling to Gi
Abstract—Galanin effects are mediated by distinct receptors, galanin receptor 1 (GAL-R1), GAL-R2 and GAL-R3. Here, we analyzed 1) the role of GAL-R1 in cholinergic transmission and peristalsis in the guinea-pig ileum using longitudinal muscle–myenteric plexus preparations and intact segments of the ileum in organ bath, and 2) the distribution of GAL-R1 immunoreactivity in the myenteric plexus with immunohistochemistry and confocal microscopy. Galanin inhibited electrically stimulated contractions of longitudinal muscle–myenteric plexus preparations with a biphasic curve. Desensitization with 1 M galanin suppressed the high potency phase of the curve, whereas the GAL-R1 antagonist, RWJ-57408 (1 M), inhibited the low potency phase. Galanin (3 M) reduced the longitudinal muscle contraction and the peak pressure, and decreased the compliance of the circular muscle. All these effects were antagonized by RWJ-57408 (1 or 10 M). RWJ-57408 (10 M) per se did not affect peristalsis parameters in normal conditions, nor when peristalsis efficiency was reduced by partial nicotinic transmission blockade with hexamethonium. In the myenteric plexus, GAL-R1 immunoreactivity was localized to neurons and to fibers projecting within the plexus and to the muscle. GAL-R1 was expressed mostly by cholinergic neurons and by some neurons containing vasoactive intestinal polypeptide or nitric oxide synthase. This study indicates that galanin inhibits cholinergic transmission to the longitudinal muscle via two separate 1
Present address: Department of Internal Medicine and Gastroenterology, University of Bologna, Bologna, Italy. *Correspondence to: C. Sternini, CURE Digestive Diseases Research Center, Building 115, Room 224, Veterans Administration Greater Los Angeles Healthcare System, 11301 Wilshire Boulevard, Los Angeles, CA 90073, USA. Tel: ⫹1-310-312-9477; fax: ⫹1-310-268-4615. E-mail address: [email protected] (C. Sternini). Abbreviations: cAMP, cyclic adenosine monophosphate; ChAT, acetylcholine transferase; FITC, fluorescein isothiocyanate; GAL-R, galanin receptor; LMMP, longitudinal muscle–myenteric plexus; MAPK, mitogen-activated protein kinase; NOS, nitric oxide synthase; PB, phosphate buffer; Pr, residual pressure; Pt, threshold pressure; Red X, Rhodamine Red-X; VIP, vasoactive intestinal polypeptide; Vr, residual volume; Vt, threshold volume.
0306-4522/04$30.00⫹0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2003.12.043
103
104
C. Sternini et al. / Neuroscience 125 (2004) 103–112
and mediates pertussis-toxin sensitive MAPK activity (Wang et al., 1998a). The main signaling pathway of GAL-R2 is Gq/11-mediated inositol phospholipid turnover and increase in intracellular calcium levels, which are not pertussis toxin sensitive (Smith et al., 1997; Wang et al., 1998a). GAL-R3 couples to Gi/o and mediates pertussistoxin sensitive inhibition of adenylate cyclase and activation of an inward K⫹ current (Smith et al., 1998). All three GAL-R mRNAs are expressed in the brain and periphery, but with differential distribution (Waters and Krause, 2000). The widespread distribution of GAL-R1 mRNA and the pharmacology corresponding to this receptor in the brain, spinal cord, gut and pancreas-derived cells suggest that this receptor mediates many of the galanin actions, including feeding, nociception, gut secretion and motility (Branchek et al., 1998). We have demonstrated that GAL-R1 immunoreactivity is expressed by enteric neurons of the rat stomach and small intestine, by fibers distributed to the musculature and mucosa, and by enterochromaffinlike cells of the stomach (Pham et al., 2002). These findings are consistent with the hypothesis that GAL-R1 mediates galanin actions on gastrointestinal motility and secretion by modulating neurotransmitter/hormone release. Since galanin acts as an inhibitory neuromodulator on cholinergic and tachykinergic transmission (Rattan, 1991; Mulholland et al., 1992), it is reasonable to hypothesize that galanin inhibits the excitatory component of peristalsis. Furthermore, the colocalization of galanin with nitric oxide and vasoactive intestinal polypeptide, the main transmitters of descending neurons, further supports the possibility that galanin affects peristalsis (Wang et al., 1998b). In the present study, we investigated the role of GAL-R1 in the galanin inhibition of cholinergic neuromuscular transmission in the longitudinal muscle–myenteric plexus (LMMP) preparation and in peristaltic activity of intact segments of the guinea-pig ileum by using a GAL-R1 specific, non-peptide antagonist, RWJ-57408 (kindly provided by Johnson Pharmaceutical Institute, Spring House, PA, USA; Scott et al., 2000). We also examined the distribution of GAL-R1 immunoreactivity in the myenteric plexus of the ileum using immunohistochemistry and confocal microscopy with a well characterized antibody raised to the third intracellular loop of GAL-R1 (GAL-R1Y225-238; Pham et al., 2002).
EXPERIMENTAL PROCEDURES Care and handling of the animals were in accordance with all NIH recommendations for the humane use of animals. All experimental procedures were reviewed and approved by the appropriate Animal Use Committee of the Institutions (UCLA and Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, USA, and the University of Pavia, Pavia, Italy) where the experiments were performed. The number of animals used was kept to the minimum necessary for a meaningful interpretation of the data and animal discomfort was kept to the minimum. Male albino, Harlan Porcellus guinea-pigs (Hartley; 250 –350 g; Harlan Laboratories, San Diego, CA, USA; G. Bettinardi; Momo, Novara, Italy) were used.
Electrically induced contractile responses Segments of the distal ileum were dissected from guinea-pig killed by CO2 inhalation and then immersed in standard Tyrode solution composed of (mM): NaCl, 136.9; KCl 2.7; CaCl2 1.8; MgCl2 1.04; NaH2PO4 0.4; NaHCO3 11.9; glucose 5.5. Five animals were used for each experimental condition. LMMP preparations were prepared by teasing the longitudinal muscle with the intact myenteric plexus from the underlying circular muscle (Tonini et al., 1998; McConalogue et al., 1999). For the analysis of electrically induced neurogenic contractions (twitch contractions), 4 cm long strips were folded in half and mounted in organ baths containing 5 ml of Tyrode solution (bubbled with 95% O2 and 5% CO2) at 37 °C, under tension of 5 mN. Isometric contractions were recorded with a force-displacement transducer (Harvard Instruments, South Natick, MA, USA). Following a 45 min equilibration period, strips were stimulated via two platinum electrodes with a Grass S44 stimulator (Grass Instruments Co., Quincy, MA, USA). Square wave pulses (0.5 ms) of supramaximal amplitude (60 V) were delivered at a frequency of 0.1 Hz. Concentration-response curves for the inhibitory effects of porcine galanin (Peninsula, Belmont, CA, USA) (1 nM–10 M), which has similar affinity for each GAL-R subtype, were calculated. The electrically evoked contractions in LMMP preparations were abolished by both tetrodotoxin (1 M) and atropine (1 M), indicating that they are mediated by activation of cholinergic nerves. Non-cumulative concentration-response curves were obtained by half-logarithmic dosing increments. The inhibitory effect of galanin on the stimulated muscle twitch was expressed as percentage of control contractions. The inhibitory effects of galanin were also determined following high (1 M) desensitizing concentrations of galanin (30 min of contact) or in the presence of the GAL-R1 non-peptide selective antagonist, RWJ-57408 (1 M prior to the application of galanin). The binding affinity IC50’s of RWJ-57408 for GAL-R1 ranged between 190 and 2700 nM in human Bowes melanoma cells (Scott et al., 2000). In addition, RWJ-57408 was effective in blocking the galanin inhibition of acetylcholine release in the CNS and the electrically induced, acetylcholine-mediated contraction in the guinea-pig ileum (Scott et al., 2000). Signals were recorded using a PowerLab data acquisition system (Analog Digital Instruments, UK) and analyzed using PowerLab Chart v4.1.1 software.
Data analysis All data are expressed as mean⫾S.E.M. Galanin inhibition curve was analyzed by fitting it to a logistic equation of the form: Effect⫽Emax/1⫹e{⫺2.303⫻slope⫻(log[A]⫺log[A50])} where, Emax⫽ maximum response; [A]⫽molar agonist concentration; [A50]⫽ molar agonist concentration inducing 50% of the maximum response. All data were fitted either to a single logistic expression or to the sum of two logistics. Goodness of fit to a single or double logistic expression was evaluated by the F test of the residual variances using a significance criterion of P⬍0.05 (SAS/STAT User’s guide, release 6.03; SAS Institute Inc., Cary, NC, USA; Lucchelli et al., 1995). Apparent affinity estimate (pKB) from a single antagonist concentration was calculated by the Gaddum (1957) equation.
Induction of peristalsis Peristalsis was studied using a previously described method (Tonini et al., 1981, 1989). Segments of ileum, approximately 7– 8 cm in length, were mounted horizontally in a 50 ml organ bath containing Tyrode solution maintained at 37 °C and gassed with 95% O2 and 5% CO2. Longitudinal movements were detected by means of an isotonic transducer loaded with 1 g. The peristalsis was elicited by delivering Tyrode solution into the lumen, by means of a peristaltic pump, at a constant rate of 0.75 ml/min. The flow was interrupted as soon as peristalsis developed, and the ejected fluid acting by gravity on a balanced switch turned off the
C. Sternini et al. / Neuroscience 125 (2004) 103–112 pump. The arrest of the inflow, controlled by a timer, was maintained for 10 s in order to release pressure on the contracted organ during propulsion, and to allow a period of rest between two peristalses. Intraluminal pressure was measured by two appropriately calibrated pressure transducers (Bell & Howell) connected to the anal end of the preparation, as an indication of circular muscle activity. One pressure transducer was calibrated at high sensitivity to record the threshold pressure for peristalsis and the other at low sensitivity to record the peak pressure (maximal ejection pressure) when the intestine was emptying. Peristaltic activity was studied in 10 min duration cycles; each period of peristaltic activity was followed by at least 15 min of rest.
105
Table 1. Characteristics of primary antibodies Antigen
Host
Dilution
Code and reference
ChAT
Goat
1:100
GAL-R1
Rabbit
1:2000–1:3000
NOS
Mouse
1:100
VIP
Mouse
1:1500
AB144P; Chemicon; Mawe et al., 1996; Li and Furness, 1998 GAL-R1Y225–238; Pham et al., 2002 N31020; Transduction Lab, Lexington, KY, USA; Williamson et al., 1996 VIP55; Wong et al., 1996
Parameters measured The following parameters were examined: a) shortening of the longitudinal muscle during the preparatory phase of peristalsis; b) peak pressure (i.e. maximum ejection pressure) that reflects the anally propagated wave of contraction of the circular muscle responsible for the clearance of the intraluminal content; c) threshold pressure (Pt); d) residual pressure (Pr); e) threshold volume (Vt); f) residual volume (Vr); g) compliance. The threshold pressure was the pressure at the beginning of peristalsis, whereas the residual pressure corresponded to the intraluminal pressure after completion of peristalsis. The threshold volume required to trigger the emptying phase was measured as the volume infused into the intestine plus the residual volume remaining at the end of the previous peristaltic wave. The residual volume, which remained in the intestine after a peristaltic wave, was estimated by deducting the expelled volume from that infused into the lumen during the preparatory (filling) phase. The compliance of the intestinal wall during the preparatory phase (⌬V/⌬P) was defined as the change in intraluminal pressure in response to a given change in intraluminal volume. This value reflects the resistance of the intestinal wall to infused fluid. It can be calculated from the inverse of the slope of the preparatory phase of the intraluminal pressure trace (Waterman et al., 1992) as it follows: average compliance⫽Vt⫺Vr/ Pt⫺Pr⫽⌬V/⌬P, where Vt⫺Vr corresponds to the volume infused during the preparatory phase. Four sets of experiments were carried out. In one set of experiments, the effects of 3 M galanin administered 5 min following the start of the cycle was evaluated. In two other sets, the effect of 1 M or 10 M RWJ-57408, administered at the start of the resting period between cycles, was assessed with and without galanin. In the fourth set of experiments, the effect of 10 M RWJ-57408 was determined in conditions of peristalsis impairment caused by 10 M hexamethonium that blocks the ganglionic nicotinic transmission (Bartho et al., 1989). Signals were recorded as described above.
Data analysis Data are expressed as mean⫾S.E.M. and represent the peristaltic activity recorded for 10 min. The effect of galanin was measured for 5 min following administration. One-way analysis of variance for repeated measures was used for statistical analysis. P⬍0.05 was considered as significant.
Immunohistochemistry Guinea-pigs were deeply anesthetized with i.p. sodium pentobarbital (Nembutal; 100 mg/kg; Abbott Laboratories, Chicago, IL, USA). The distal ileum was dissected, opened up along the mesentery, pinned flat on wax and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4 for 2 h at room temperature and then stored in PB with 0.1% g sodium azide at 4 °C before processing for immunohistochemistry as whole mounts (Sternini et al., 2000). For colchicine treatment, a segment of the distal ileum, pinned flat as described above, was placed in sterile Krebs solution (in mM: KCl, 5.9; NaCl, 118; CaCl2, 2.5; MgSO4, 1.2;
NaH2PO4, 1.4; glucose, 5; Fungizone, 2.5 g/ml; penicillin, 100 IU/ml; and streptomycin, 100 g/ml), bubbled with 95% O2:5% CO2, pH 7.4, at 37 °C, washed and then incubated in Dulbecco’s modified Eagle’s medium (DME/F12) containing 10% v/v fetal bovine serum, penicillin, streptomycin and Fungizone at 37 °C with 95% O2:5% CO2, pH 7.4, for 30 min, transferred to medium containing colchicine (100 M; Sigma Chemical, St. Louis, MO, USA) for 16 –20 h at 37 °C, and then fixed as described above. Tissues were processed by the immunofluorescence method for single and double labeling (Sternini et al., 2000; Pham et al., 2002). Briefly, tissues were washed in PB, incubated in 10% normal donkey serum for 1 h at room temperature to minimize the background, incubated in GAL-R1 antiserum (1:2000 –1:3000) for 3 days at 4 °C, followed by a 2-h incubation in affinity purified donkey anti-rabbit IgG conjugated with fluorescein isothiocyanate (FITC; Jackson Immunolabs, West Grove, PA, USA) at 1:100 dilution, washed and mounted on slides and then coverslipped. The GAL-R1 antiserum has been characterized by immunostaining and Western blot of membrane from cells transfected with GAL-R1 cDNA and from control cells (Pham et al., 2002). For double-label immunofluorescence, tissues were incubated in a mixture of rabbit GAL-R1 antiserum and mouse monoclonal antibody to vasoactive intestinal polypeptide (VIP) at 1:1500 dilution or nitric oxide synthase (NOS; Chemicon, Temecula, CA, USA) at 1:50 dilution; or goat acetylcholine transferase (ChAT; Chemicon) at 1:100 dilution for 3 days at 4 °C, followed by a 2-h incubation at room temperature with a mixture of affinity-purified donkey anti-rabbit IgG coupled with FITC (1:100; Jackson Immunolabs) and donkey antimouse or anti-goat IgG conjugated with Rhodamine Red-X (Red-X; 1:300 –1:500; Jackson Immunolabs), then coverslipped with 90% glycerol containing 2% potassium iodide in 0.1 M PB. Primary and secondary antibodies were diluted in 0.5% Triton X-100 in 0.1 M PB. Specificity controls included preadsorption of the GAL-R1 with the peptide used for immunization (10 M) for 16 –24 h at 4 °C. ChAT, VIP and NOS antibodies (Table 1) have been previously validated for immunohistochemical use in gut tissue (Mawe et al., 1996; Williamson et al., 1996; Wong et al., 1996; Li and Furness, 1998). Three colchicine-treated guinea-pigs were used for cell counting. For each animal, 10 plexuses were randomly selected, the number of myenteric neurons expressing GAL-R1, ChAT, VIP or NOS were counted, and the mean percentage of GAL-R1 neurons containing each of the other marker was calculated. Tissues were examined using a light Leitz Orthoplan Microscope using a Phloem illumination system with “L2” and “N2” filters for FITC and Red-X visualization, and a dualband filter for the simultaneous visualization of FITC and Red-X. A Zeiss laser scanning microscope 410 with a krypton/argon laser and attached to a Zeiss Axiovert 100 microscope with PlanApo 63⫻1.3 na objective was also used. Typically, 10 –20 optical sections were taken with a z-axis of 0.75 m. Images were processed and labeled
106
C. Sternini et al. / Neuroscience 125 (2004) 103–112
RWJ-57408 (1 M), caused a parallel shift to the right of the low potency phase of the galanin curve, without affecting the high potency phase (Fig. 3B). The apparent affinity estimate (pKB) of RWJ-57408 was 6.6. Effects of galanin on peristalsis
Fig. 1. Inhibitory effect of galanin on electrically stimulated twitch contractions of LMMP preparations. Galanin inhibition curve was biphasic. The potency (pEC50(H)) of the first phase was 8.2, while that of the second phase (pEC50(L)) was 6.6. The high potency phase (in the nanomolar concentration range) reached a maximum of about 25%. Each point represents the mean⫾S.E.M. (n⫽5).
using Adobe Photoshop 6.0 (Adobe Systems, Mountain View, CA, USA).
RESULTS Effects of galanin on twitch contractions in LMMP preparations Galanin (1 nM–10 M) induced a concentration-dependent inhibition of electrically stimulated muscle twitch contractions (Fig. 1) up the their abolition. The curve to galanin was better fitted to a biphasic than to a monophasic model (F⫽45.63; P⬍0.001). The potency estimate of the first phase (high potency phase: pEC50(H)) was 8.2, while that of the second low potency phase (pEC50(L)) was 6.6. In unstimulated LMMP preparations, submaximal contraction to acetylcholine (0.1 M) was not affected by 1 M galanin (control: 10.5⫾1.0 mN vs. 9.7⫾1.3 mN in the presence of galanin, n⫽5; Fig. 2). As shown in the tracing, galanin caused a slight contractile response (about 1.5 mN in amplitude). This response and the contractile response produced by acetylcholine were not modified by 1 M tetrodotoxin, a blocker of voltage-dependent neuronal Na⫹-channels (n⫽3), indicating a direct effect on the effector cells, with the galanin contraction probably being mediated by a postjunctional excitatory receptor associated with intracellular calcium mobilization. To desensitize GAL-R, a high (1 M) concentration of galanin was used and left in the bath for 30 min. The initial inhibitory effect of galanin on twitch contractions (approximately 70%) gradually vanished with a recovery of control twitch amplitude by 90% indicating receptor desensitization. Under these conditions, the high potency phase of the galanin inhibitory curve was completely suppressed and the curve became monophasic. The low potency phase was also reduced without suppression of the maximum response (Fig. 3A). By contrast, the GAL-R1 antagonist,
Intraluminal fluid infusion produced the two phases of peristalsis, a preparatory phase followed by an emptying phase. During the preparatory phase, there was a slight and slow increase of the intraluminal pressure due to the accommodation processes of the circular muscle (Waterman et al., 1994a). At the same time, there was a shortening of the longitudinal muscle, which reached the maximum when the emptying phase occurred. This phase was characterized by a threshold pressure of approximately 100 Pa, a threshold volume of about 900 l, and a peak pressure (maximum ejection pressure) of about 1800 Pa. The frequency of peristaltic waves in a 10 min-cycle was 8.4⫾0.4 (n⫽11; Fig. 4). Galanin (3 M) significantly reduced the longitudinal muscle contraction (P⬍0.05) during the preparatory phase of peristalsis (Fig. 5) and the peak pressure (P⬍0.002; Fig. 6) generated during the emptying phase. The inhibitory effect of galanin on longitudinal muscle contractions and peak pressure was antagonized by RWJ-57408 (1 and 10 M). RWJ-57408 per se did not have any effect on these nor on the following parameters (not shown). Galanin (3 M) did not affect the threshold volume (Fig. 7) and pressure (Fig. 8) required to trigger the emptying phase of peristalsis, but significantly increased the residual volume (P⬍0.002) and pressure (P⬍0.05) after completion of a peristaltic event. RWJ-57408 antagonized these galanin effects at 10 M, but not at 1 M (Figs. 7, 8). Galanin (3 M) decreased the compliance of the intestinal wall by 25%. This effect was antagonized by the GAL-R1 antagonist at 10 M (P⬍0.06 vs. control; Fig. 9). The lack of effect of RWJ-57408 per se on peristalsis casts doubt on the
Fig. 2. Effect of galanin on exogenously applied acetylcholine (ACh: 0.1 M) in resting LMMP preparations. Note that a concentration of galanin (1 M) that induces ⬎50% inhibition of twitch contractions does not affect the contractile response induced by exogenous acetylcholine. Dots indicate drug administration.
C. Sternini et al. / Neuroscience 125 (2004) 103–112
107
Fig. 3. (A) Inhibitory effect of galanin on electrically stimulated contractions of LMMP preparations (open circle) or following self-desensitization with a high concentration (1 M for 30 min) of galanin (closed triangle). Galanin inhibited the amplitude of twitch contractions up to their abolition. Note that the desensitizing concentration of galanin completely suppressed the high potency phase of galanin curve, and partially reduced the low potency phase without abolishing its maximum. (B) Galanin inhibitory curve in the presence (closed circle) and absence (open circle) of the GAL-R1 antagonist (RWJ-57408: 1 M). Note that the GAL-R1 antagonist inhibits the low potency phase of galanin curve, whereas it did not affect the high potency phase. Values are expressed as mean⫾S.E.M. (n⫽5).
functional role of endogenous galanin in modulating propulsive activity. On the other hand, a subtle tuning of
neuronal activity cannot be detected with an intact cholinergic transmission, which is the main excitatory trans-
Fig. 4. Recording of peristalsis in the guinea-pig isolated ileum during a 10 min cycle. The top tracing represents the shortening (mm) of the longitudinal muscle, whereas the middle and bottom tracings illustrate the intraluminal pressure (Pascal: Pa) recorded at high and low sensitivity, respectively. Arrows indicate the threshold (Pt) and residual (Pr) pressure (middle tracing) and the peak pressure (bottom tracing). The peak pressure in the middle tracing is truncated because it is off scale.
108
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Fig. 5. Effect of galanin (Gal: 3 M) on longitudinal muscle shortening (mm) during the preparatory (filling) phase of peristalsis in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates longitudinal muscle shortening in the absence of drug treatment. Galanin significantly inhibits the shortening of longitudinal muscle through a mechanism antagonized by RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
mission regulating peristalsis. To overcome this problem, a series of experiments were conducted in the presence of the ganglionic blocking agent, hexamethonium. Hexamethonium induced an increase of the threshold and residual volume at the end of the preparatory phase of peristalsis (Fig. 10), which are considered a representative measure of peristalsis impairment (Waterman et al., 1992). Even under these conditions of partial blockade of cholinergic transmission, which should facilitate the detection of small modulatory effects of endogenous galanin on GAL-R1, RWJ-57408 at 10 M did not have any effect (Fig. 10).
Fig. 6. Effect of galanin (Gal: 3 M) on peak pressure (Pascal: Pa) of each peristaltic wave in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates pressure values in the absence of drug treatment. Galanin significantly inhibits the peak pressure through a mechanism antagonized by RWJ-57408 (10 M). Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control. Note that RWJ-57408 treatment (10 M) in the presence of galanin causes a response, which is not different from control response indicating a full antagonism of galanin inhibitory effect.
Fig. 7. Effect of galanin (Gal: 3 M) on the threshold (Vt) and residual (Vr) volume (l) in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates volume values in the absence of drug treatment. Galanin did not affect the threshold volume, but significantly increased the residual volume through a mechanism antagonized by 10 M RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
GAL-R1 immunoreactivity in the myenteric plexus of the guinea-pig ileum GAL-R1 immunoreactivity was observed in numerous neurons of the myenteric plexus (Fig. 11), and in fibers within the plexuses and distributed to the muscle layer (not shown) with a pattern similar to the one described in the rat gastrointestinal tract (Pham et al., 2002). Confocal microscopy was required to discern cell bodies due to the dense neuropil formed by GAL-R1 immunoreactive fibers in the myenteric plexus. Double labeling studies revealed that the vast majority (about 80%) of GAL-R1 myenteric neurons contained ChAT immunoreactivity (Fig. 11). However, there were also GAL-R1 myenteric neurons immunoreactive for VIP or NOS (about 12% and 18%, respectively; Fig. 11).
Fig. 8. Effect of galanin (Gal: 3 M) on the threshold (Pt) and residual (Pr) pressure (Pascal: Pa) in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates pressure values in the absence of drug treatment. Galanin did not affect the threshold pressure, but significantly increased the residual pressure through a mechanism antagonized by 10 M RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Fig. 9. Effect of galanin (Gal: 3 M) on circular muscle compliance (L/Pa) in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates compliance values in the absence of drug treatment. Galanin significantly reduces the compliance through a mechanism antagonized by 10 M RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
DISCUSSION These studies demonstrate that 1) galanin inhibits cholinergic transmission to the longitudinal muscle by acting at a prejunctional level via the activation of at least two receptors, one being GAL-R1, since the low potency phase of the galanin inhibitory curve is antagonized by the GAL-R1 antagonist, RWJ-57408; 2) galanin affects peristalsis by reducing its efficiency (reduction of longitudinal muscle contraction and peak pressure) and decreasing the compliance of the intestinal wall through activation of GAL-R1; and 3) GAL-R1 immunoreactivity is localized to functionally distinct neurons of both the ascending and descending pathways. Galanin effect on neurogenic twitch contractions was prejunctional, since submaximal contractions to exogenous acetylcholine in unstimulated LMMP preparations were not
Fig. 10. Effect of hexamethonium (Hex: 10 M) on threshold and residual volume. Control C indicates volume in the absence of any treatment. Note that the significant increase of threshold and residual volume induced by hexamethonium is not affected by RWJ-57408 (RWJ) at 10 M (n⫽8). * P⬍0.05 vs. control.
109
affected. The inhibitory curve caused by galanin was biphasic, with the low potency phase antagonized by RWJ-57408 and the high potency phase abolished by GAL-R desensitization. This is consistent with the involvement of two receptors/mechanisms including the GAL-R1. It is likely that the high potency phase (approximately 25% of the total curve) is mediated by the GAL-R3, an inhibitory receptor, which utilizes the same transduction mechanisms as the GAL-R1 (Smith et al., 1998), although selective antagonists at the GAL-R3 are required to clarify this point. Indeed, GAL-R1 and GAL-R3 mediate pertussis toxin sensitive inhibition of adenylate cyclase. The other receptor activated by galanin, the GAL-R2, is an excitatory receptor mainly coupled to Gq/11 signaling pathway (Smith et al., 1997; Wang et al., 1998a). This receptor induces direct muscle contraction and transmitter release; therefore it cannot be responsible for the inhibitory effect caused by nanomolar concentrations of galanin. It has been reported that galanin inhibits electrically induced neurogenic cholinergic contractions and acetylcholine release in the guinea-pig ileum (Ekblad et al., 1985a; Yau et al., 1986; Mulholland et al., 1992; Akehira et al., 1995). Our findings extend previous results, indicating that the inhibitory effect of galanin is mediated by two distinct receptors, which undergo different degree of desensitization. The morphological data showing that the vast majority of GAL-R1 neurons are cholinergic support the conclusion that this receptor plays a major role in the galanin inhibition of excitatory neuromuscular transmission. Peristalsis is a polarized motor event that propels the intestinal contents anally and depends entirely on enteric neurons (Kosterlitz and Lees, 1964; Tonini et al., 1981; Waterman et al., 1992). Distension of the gut or mucosal stimulation induces oral reflex contraction and aboral relaxation. These reflexes are initiated by sensory neurons, which make contact with ascending and descending neurons (Waterman et al., 1994b). In the isolated small intestine, peristalsis consists of a preparatory phase during which intraluminal fluid infusion provides a radial mechanical stimulus to the circular muscle up to a threshold volume that triggers the emptying phase. This phase initiates at the most oral end of the preparation with an anally propagating circular muscle contraction causing fluid displacement (Tonini et al., 1981; Waterman et al., 1992, 1994b). Galanin reduces peristaltic efficiency in the guinea-pig ileum by affecting several parameters of peristalsis. It inhibits the longitudinal muscle shortening and the peak pressure through a pathway involving GAL-R1 as indicated by the RWJ-57408 antagonism. Since shortening of longitudinal muscle and peak pressure are mediated by excitatory neurons, the observations that they are inhibited by RWJ-57408 are in agreement with our findings with LMMPs supporting a major role of GAL-R1 in the galanin inhibition of excitatory neuromuscular transmission. In addition to a reduction of the longitudinal muscle contraction and peak pressure, galanin induced an increase of the residual volume and pressure, which indicates that part of the intraluminal content of the intestine cannot be expelled. Both these parameters are antagonized by 10 M RWJ-
110
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Fig. 11. Confocal images of myenteric plexuses of the guinea-pig ileum showing GAL-R1 at the cell surface (green fluorescence), ChAT, VIP and NOS (red fluorescence) immunoreactivity in the cytoplasm of myenteric neurons. GAL-R1&ChAT, GAL-R1&VIP, and GAL-R1&NOS: simultaneous visualization of GAL-R1 and ChAT, VIP or NOS immunoreactivity. Arrows point to examples of neurons that contain GAL-R1 and ChAT, VIP or NOS immunoreactivity. Scale bars: GAL-R1&ChAT⫽10 m; GAL-R1&VIP⫽20 m; GAL-R1&NOS⫽15 m.
57408. This suggests that GAL-R1 participates in the mechanisms responsible for the intraluminal clearance. Galanin decreases the compliance of the circular muscle during the preparatory phase of peristalsis. Compliance reflects the resistance of the intestinal wall to the infused intraluminal fluid and it is controlled by nonadrenergic, noncholinergic inhibitory (nitrergic) pathways (Waterman et al., 1994a). The inhibitory effect of galanin on compliance was affected by RWJ-57408, suggesting the involvement of GAL-R1. This is in agreement with the presence of GAL-R1 on descending myenteric neurons, as shown by the colocalization of GAL-R1 with VIP and NOS immunoreactivity. However, we cannot exclude the participation of an excitatory receptor (like the GAL-R2) located on smooth muscle cells, which may partly hinder the accommodation of the circular muscle during the preparatory phase of peristalsis. The failure of RWJ-57408 to affect peristalsis per se suggests that endogenous galanin does not exert a functional tonic inhibition on neuronal pathways subserving
peristalsis through GAL-R1. However, the inhibitory action of galanin could be undetectable when excitatory cholinergic transmission is intact. The lack of effect of RWJ57408 on peristalsis when cholinergic nicotinic transmission is partially blocked by hexamethonium excludes a modulatory role of endogenous galanin on neuronal pathways controlling peristalsis. This implies that in our conditions, galanin exerts a pharmacological rather than a physiological role on peristaltic activity. GAL-R1 is localized to functionally distinct populations of enteric neurons and to fibers distributed to the muscle. This pattern is comparable to the one we have described for the rat stomach and small intestine (Pham et al., 2002). The distribution of GAL-R1 neurons and their sites of innervation are in accord with the galanin distribution (Ekblad et al., 1985b; Burgevin et al., 1995; Pham et al., 2002). The presence of GAL-R1 on neuronal cell surface and processes distributed to the muscle is consistent with pre- and post-synaptic actions of galanin via this receptor and supports the concept that GAL-R1 is transported to
C. Sternini et al. / Neuroscience 125 (2004) 103–112
axon terminals. In the enteric nervous system, neurons can be differentiated on the basis of their morphology and projections, and of the combination of transmitters that they synthesize and release (Costa et al., 1996). The finding that as many as 80% of GAL-R1 myenteric neurons contain ChAT supports the concept that GAL-R1 bearing myenteric neurons comprise ascending excitatory neurons. This is in agreement with the functional data showing that galanin inhibits the contractile response to electrical stimulation (this paper; Ekblad et al., 1985a; Yau et al., 1986; Mulholland et al., 1992; Akehira et al., 1995) through the inhibition of excitatory transmitters like acetylcholine and substance P via GAL-R1. The observation that the inhibitory effect of galanin on stimulated acetylcholine release is reversed by pertussis toxin further supports the involvement of GAL-R1. However, the presence of GAL-R1 on VIP and NOS immunoreactive neurons, in addition to ChAT containing neurons, is consistent with a modulatory role of this receptor on both the ascending and descending pathways. In summary, on the basis of our functional and morphological studies, we can conclude that galanin effects on neuromuscular transmission in the longitudinal muscle and on peristaltic activity in the guinea-pig ileum involve the activation of GAL-R1 on ascending and descending pathways. The failure of RWJ-57408 to modify per se cholinergic neuromuscular transmission and peristalsis, despite a redundant presence of GAL-R1 in myenteric neurons, suggests that under our experimental conditions endogenous galanin does not exert a tonically active modulatory role on gut motility via GAL-R1. Acknowledgements—This paper was supported by National Institutes of Health grants, DKs 57037 (to C.S.), 35740 (to C.S.), 41301 (CURE:DDRCC: Morphology/Imaging Core to C.S.), and University of Pavia, Italy Funds (to M.T.).
REFERENCES Akehira K, Nakane Y, Hioki K, Taniyama K (1995) Site of action of galanin in the cholinergic transmission of guinea pig small intestine. Eur J Pharmacol 284:149 –155. Bartfai T, Fisone G, Langel U (1992) Galanin and galanin antagonists: molecular and biochemical perspectives. Trends Pharmacol Sci 13:312–317. Bartho L, Holzer P, Leander S, Lembeck F (1989) Evidence for an involvement of substance P, but not cholecystokinin-like peptides, in hexamethonium-resistant intestinal peristalsis. Neuroscience 28: 211–217. Bauer FE, Zintel A, Kenny MJ, Calder D, Ghatei MA, Bloom SR (1989) Inhibitory effect of galanin on postprandial gastrointestinal motility and gut hormone release in humans. Gastroenterology 97:260 – 264. Bedecs K, Berthold M, Bartfai T (1995) Galanin-10 years with a neuroendocrine peptide. Int J Biochem Cell Biol 27:337–349. Branchek T, Smith KE, Walker MW (1998) Molecular biology and pharmacology of galanin receptors. Ann NY Acad Sci 863:94 –107. Branchek TA, Smith KE, Gerald C, Walker MW (2000) Galanin receptor subtypes. Trends Pharmacol Sci 21:109 –117. Burgevin MC, Loquet I, Quarteronet D, Habert-Ortoli E (1995) Cloning, pharmacological characterization, and anatomical distribution of a rat cDNA encoding for a galanin receptor. J Mol Neurosci 6:33–41. Costa M, Brookes SJH, Steele PA, Gibbins I, Burcher E, Kandiah CJ
111
(1996) Neurochemical classification of myenteric neurons in the guinea pig ileum. Neuroscience 75:949 –967. Ekblad E, Hakanson R, Sundler F, Walhestedt C (1985a) Galanin: neuromodulatory and direct contractile effects on smooth muscle preparations. Br J Pharmacol 86:241–246. Ekblad E, Rokaeus A, Hakanson R, Sundler F (1985b) Galanin nerve fibers in the rat gut: distribution, origin and projections. Neuroscience 16:355–363. Floren A, Land T, Langel U (2000) Galanin receptor subtypes and ligand binding. Neuropeptides 34:331–337. Fox JET, McDonald TJ, Kostolanska F, Tatemotot K (1986) Galanin: an inhibitory neural peptide of the canine small intestine. Life Sci 39:103–110. Furness JB, Costa M, Rokaeus A, McDonald TJ, Brooks B (1987) Galanin-immunoreactive neurons in the guinea-pig small intestine: their projections and relationships to other enteric neurons. Cell Tissue Res 250:607–615. Gaddum JH (1957) Series of drug antagonism. Pharmacol Rev 9:211– 218. Gu ZF, Pradhan TK, Coy DH, Jensen RT (1995) Interaction of galanin fragments with galanin receptors on isolated smooth muscle cells from guinea pig stomach: identification of a novel galanin receptor subtype. J Pharmacol Exp Ther 272:371–378. Kosterlitz HW, Lees GM (1964) Pharmacological analysis of intrinsic intestinal reflexes. Pharmacol Rev 16:301–339. Li ZS, Furness JH (1998) The immunohistochemical localization of cholinergic markers in putative intrinsic primary afferent neurons of the guinea pig small intestine. Cell Tissue Res 294:35–43. Lucchelli A, Santagostino-Barbone MG, Barbieri A, Candura SM, Tonini M (1995) The interaction of antidepressant drugs with central and peripheral (enteric) 5-HT3 and 5-HT4 receptors. Br J Pharmacol 114:117–125. Mawe GM, Talmage EK, Lee KP, Parsons RL (1996) Expression of choline acetyltransferase immunoreactivity in guinea pig cardiac ganglia. Cell Tissue Res 285:281–286. McConalogue K, Grady EF, Minnis J, Balestra B, Tonini M, Brecha NC, Bunnett NW, Sternini C (1999) Activation and internalization of the opioid receptor by the newly described endogenous agonists, endomorphin-1 and endomorphin-2. Neuroscience 90:1051–1059. Melander T, Hokfelt T, Rokaeus A, Fahrenkrug J, Tatemoto K, Mutt V (1985) Distribution of galanin-like immunoreactivity in the gastrointestinal tract of several mammalian species. Cell Tissue Res 239:253–270. Mulholland MW, Schoeneich S, Flowe K (1992) Galanin inhibition of enteric cholinergic neurotransmission: guanosine triphosphatebinding protein interactions with adenylate cyclase. Surgery 112: 195–201. Muramatsu I, Yanaihara N (1988) Contribution of galanin to noncholinergic, non-adrenergic transmission in rat ileum. Br J Pharmacol 94:1241–1249. Palmer JM, Schemmann M, Tamura K, Wood JD (1986) Galanin mimics slow synaptic inhibition in myenteric neurons. Eur J Pharmacol 124:379 –380. Pham T, Guerrini S, Wong E, Reeve JR Jr, Sternini C (2002) Distribution of galanin receptor 1 immunoreactivity in the rat stomach and small intestine. J Comp Neurol 450:292–302. Rattan S (1991) Role of galanin in the gut. Gastroenterology 100: 1762–1768. Scott MK, Ross TM, Lee DH, Wang HY, Shank RP, Wild KD, Davis CB, Crooke JJ, Potocki AC, Reitz AB (2000) 2,3-Dihydro-dithiin and -dithiepine-1,1,4,4-tetroxides: small molecule non-peptide antagonists of the human galanin hGAL-1 receptor. Bioorg Med Chem 8:1383–1391. Smith KE, Forray C, Walker MW, Jones KA, Tamm JA, Bard J, Branchek TA, Linemeyer DL, Gerald C (1997) Expression cloning of a rat hypothalamic galanin receptor coupled to phosphoinositide turnover. J Biol Chem 272:24612–24616. Smith KE, Walker MW, Artymyshyn R, Bard J, Borowsky B, Tamm JA,
112
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Yao W-J, Vaysse PJ-J, Branchek TA, Gerald C, Jones KA (1998) Galanin GALR3 receptors: cloning and functional expression of rat and human receptors. J Biol Chem 273:233321–233326. Sternini C, Brecha NC, Minnis J, D’Agostino G, Balestra B, Fiori E, Tonini M (2000) Role of agonist-dependent receptor internalization in the regulation of opioid receptors. Neuroscience 98:233–241. Tamura K, Palmer JM, Wood JD (1987) Galanin suppresses nicotinic synaptic transmission in the myenteric plexus of the guinea pig ileum. Eur J Parmacol 136:445–446. Tonini M, Fiori E, Balestra B, Spelta V, D’Agostino G, Di Nucci A, Brecha N, Sternini C (1998) Endomorphin-1 and endomorphin-2 activate opioid receptors in myenteric neurons of the guinea-pig small intestine. Naunyn Schmiedebergs Arch Pharmacol 358:686–689. Tonini M, Frigo G, Lecchini S, D’Angelo L, Crema A (1981) Hyoscineresistant peristalsis in guinea-pig ileum. Eur J Parmacol 71:375–381. Tonini M, Galligan JJ, North RA (1989) Effects of cisapride on cholinergic neurotransmission and propulsive motility in the guinea pig ileum. Gastroenterology 96:1257–1264. Wang S, Hashemi T, Fried S, Clemmons AL, Hawes BE (1998a) Differential intracellular signaling of the GalR1 and GalR2 galanin receptor subtypes. Biochemistry 37:6711–6717. Wang YF, Mao YK, Fox-Threlkeld JE, McDonald TJ, Daniel EE (1998b) Colocalization of inhibitory mediators, NO, VIP and galanin, in canine enteric nerves. Peptides 19:99 –112.
Waterman SA, Costa M, Tonini M (1992) Modulation of peristalsis in the guinea-pig isolated small intestine by exogenous and endogenous opioids. Br J Pharmacol 106:1004 –1010. Waterman SA, Costa M, Tonini M (1994a) Accommodation mediated by enteric inhibitory reflexes in the isolated guinea-pig small intestine. J Physiol (Lond) 474:539 –546. Waterman SA, Tonini M, Costa M (1994b) The role of ascending excitatory and descending inhibitory pathways in peristalsis in the isolated guinea-pig small intestine. J Physiol (Lond) 481:223–232. Waters SM, Krause JE (2000) Distribution of galanin-1, -2 and -3 receptor messenger RNAs in central and peripheral tissues. Neuroscience 95:265–271. Williamson S, Pompolo S, Furness JB (1996) GABA and nitric oxide synthase immunoreactivities are colocalized in a subset of inhibitory motor neurons of the guinea pig small intestine. Cell Tissue Res 284:29 –37. Wong HC, Sternini C, Lloyd K, De Giorgio R, Walsh JH (1996) Monoclonal antibody to VIP: production, characterization, immunoneutralizing activity, and usefulness in cytochemical staining. Hybridoma 15:133–139. Yau WM, Dorsett JA, Youther ML (1986) Evidence for galanin as an inhibitory neuropeptide on myenteric cholinergic neurons in the guinea pig small intestine. Neurosci Lett 72:305–308.
(Accepted 18 December 2003)
ROLE OF GALANIN RECEPTOR 1 IN PERISTALTIC ACTIVITY IN THE GUINEA PIG ILEUM C. STERNINI,a,b* L. ANSELMI,a,b,c S. GUERRINI,a,b1 E. CERVIO,c T. PHAM,a,b B. BALESTRA,c R. VICINI,c P. BAIARDI,d G.-L. D’AGOSTINOe AND M. TONINIc
receptors; GAL-R1 mediates the low potency phase. The reduced peristalsis efficiency could be explained by inhibition of the cholinergic drive, whereas the decreased compliance is probably due to inhibition of descending neurons and/or to the activation of an excitatory muscular receptor. Endogenous galanin does not appear to affect neuronal pathways subserving peristalsis in physiologic conditions via GAL-R1. © 2004 IBRO. Published by Elsevier Ltd. All rights reserved.
a CURE Digestive Diseases Research Center, Digestive Diseases Division, Building 115, Room 224, Veterans Administration Greater Los Angeles Healthcare System, 11301 Wilshire Boulevard, Los Angeles, CA 90073, USA b Departments of Medicine and Neurobiology, University of California Los Angeles, David Geffen School of Medicine, Los Angeles, CA 90095, USA
Key words: cholinergic transmission, descending myenteric neurons, ascending myenteric neurons, longitudinal muscle– myenteric plexus preparations, noncholinergic, nonadrenergic inhibitory pathways.
c
Department of Physiological and Pharmacological Sciences, University of Pavia, Pavia, Italy
d
S. Maugeri Foundation, University of Pavia, Pavia, Italy
e
Department of Experimental and Applied Pharmacology, University of Pavia, Pavia, Italy
Galanin is a brain– gut neuropeptide that exerts a variety of cellular functions in the nervous system, including neurotransmitter and hormone release, nociception, spinal reflexes and feeding behavior (Rattan, 1991; Bartfai et al., 1992; Bedecs et al., 1995). In the gastrointestinal tract, galanin is localized to myenteric and submucosal neurons and to fibers projecting to the gut wall (Ekblad et al., 1985b; Melander et al., 1985; Furness et al., 1987), and it affects motility and secretion mainly acting as an inhibitory neuromodulator (Rattan, 1991). The effects of galanin on gastrointestinal smooth muscle are both stimulatory and inhibitory and are the result of a direct myogenic effect and a nerve-mediated effect involving the release of other transmitters (Ekblad et al., 1985a; Fox et al., 1986; Yau et al., 1986; Muramatsu and Yanaihara, 1988; Bauer et al., 1989; Mulholland et al., 1992; Akehira et al., 1995; Gu et al., 1995). For instance, in the guinea-pig intestine, galanin inhibits the electrically induced contraction mediated by endogenous acetylcholine and substance P, and acetylcholine release (Ekblad et al., 1985a; Fox et al., 1986; Yau et al., 1986; Muramatsu and Yanaihara, 1988; Bauer et al., 1989; Mulholland et al., 1992; Akehira et al., 1995). Galanin also hyperpolarizes myenteric neurons and suppresses their excitability (Palmer et al., 1986; Tamura et al., 1987). Together, these pharmacological studies indicate that galanin inhibits the excitatory neuroneuronal and neuromuscular transmission in the guinea-pig ileum, and it exerts direct effects on smooth muscle cells. Galanin effects are mediated by distinct receptors that activate multiple second messenger pathways to affect cell activity (Branchek et al., 1998, 2000; Floren et al., 2000). Three G protein-coupled receptors for galanin have been cloned, named galanin receptor 1 (GAL-R1), GAL-R2 and GAL-R3, which have distinct pharmacological profiles and activate different intracellular signaling pathways (Branchek et al., 2000; Floren et al., 2000). GAL-R1 inhibits forskolin-stimulated cAMP production by coupling to Gi
Abstract—Galanin effects are mediated by distinct receptors, galanin receptor 1 (GAL-R1), GAL-R2 and GAL-R3. Here, we analyzed 1) the role of GAL-R1 in cholinergic transmission and peristalsis in the guinea-pig ileum using longitudinal muscle–myenteric plexus preparations and intact segments of the ileum in organ bath, and 2) the distribution of GAL-R1 immunoreactivity in the myenteric plexus with immunohistochemistry and confocal microscopy. Galanin inhibited electrically stimulated contractions of longitudinal muscle–myenteric plexus preparations with a biphasic curve. Desensitization with 1 M galanin suppressed the high potency phase of the curve, whereas the GAL-R1 antagonist, RWJ-57408 (1 M), inhibited the low potency phase. Galanin (3 M) reduced the longitudinal muscle contraction and the peak pressure, and decreased the compliance of the circular muscle. All these effects were antagonized by RWJ-57408 (1 or 10 M). RWJ-57408 (10 M) per se did not affect peristalsis parameters in normal conditions, nor when peristalsis efficiency was reduced by partial nicotinic transmission blockade with hexamethonium. In the myenteric plexus, GAL-R1 immunoreactivity was localized to neurons and to fibers projecting within the plexus and to the muscle. GAL-R1 was expressed mostly by cholinergic neurons and by some neurons containing vasoactive intestinal polypeptide or nitric oxide synthase. This study indicates that galanin inhibits cholinergic transmission to the longitudinal muscle via two separate 1
Present address: Department of Internal Medicine and Gastroenterology, University of Bologna, Bologna, Italy. *Correspondence to: C. Sternini, CURE Digestive Diseases Research Center, Building 115, Room 224, Veterans Administration Greater Los Angeles Healthcare System, 11301 Wilshire Boulevard, Los Angeles, CA 90073, USA. Tel: ⫹1-310-312-9477; fax: ⫹1-310-268-4615. E-mail address: [email protected] (C. Sternini). Abbreviations: cAMP, cyclic adenosine monophosphate; ChAT, acetylcholine transferase; FITC, fluorescein isothiocyanate; GAL-R, galanin receptor; LMMP, longitudinal muscle–myenteric plexus; MAPK, mitogen-activated protein kinase; NOS, nitric oxide synthase; PB, phosphate buffer; Pr, residual pressure; Pt, threshold pressure; Red X, Rhodamine Red-X; VIP, vasoactive intestinal polypeptide; Vr, residual volume; Vt, threshold volume.
0306-4522/04$30.00⫹0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2003.12.043
103
104
C. Sternini et al. / Neuroscience 125 (2004) 103–112
and mediates pertussis-toxin sensitive MAPK activity (Wang et al., 1998a). The main signaling pathway of GAL-R2 is Gq/11-mediated inositol phospholipid turnover and increase in intracellular calcium levels, which are not pertussis toxin sensitive (Smith et al., 1997; Wang et al., 1998a). GAL-R3 couples to Gi/o and mediates pertussistoxin sensitive inhibition of adenylate cyclase and activation of an inward K⫹ current (Smith et al., 1998). All three GAL-R mRNAs are expressed in the brain and periphery, but with differential distribution (Waters and Krause, 2000). The widespread distribution of GAL-R1 mRNA and the pharmacology corresponding to this receptor in the brain, spinal cord, gut and pancreas-derived cells suggest that this receptor mediates many of the galanin actions, including feeding, nociception, gut secretion and motility (Branchek et al., 1998). We have demonstrated that GAL-R1 immunoreactivity is expressed by enteric neurons of the rat stomach and small intestine, by fibers distributed to the musculature and mucosa, and by enterochromaffinlike cells of the stomach (Pham et al., 2002). These findings are consistent with the hypothesis that GAL-R1 mediates galanin actions on gastrointestinal motility and secretion by modulating neurotransmitter/hormone release. Since galanin acts as an inhibitory neuromodulator on cholinergic and tachykinergic transmission (Rattan, 1991; Mulholland et al., 1992), it is reasonable to hypothesize that galanin inhibits the excitatory component of peristalsis. Furthermore, the colocalization of galanin with nitric oxide and vasoactive intestinal polypeptide, the main transmitters of descending neurons, further supports the possibility that galanin affects peristalsis (Wang et al., 1998b). In the present study, we investigated the role of GAL-R1 in the galanin inhibition of cholinergic neuromuscular transmission in the longitudinal muscle–myenteric plexus (LMMP) preparation and in peristaltic activity of intact segments of the guinea-pig ileum by using a GAL-R1 specific, non-peptide antagonist, RWJ-57408 (kindly provided by Johnson Pharmaceutical Institute, Spring House, PA, USA; Scott et al., 2000). We also examined the distribution of GAL-R1 immunoreactivity in the myenteric plexus of the ileum using immunohistochemistry and confocal microscopy with a well characterized antibody raised to the third intracellular loop of GAL-R1 (GAL-R1Y225-238; Pham et al., 2002).
EXPERIMENTAL PROCEDURES Care and handling of the animals were in accordance with all NIH recommendations for the humane use of animals. All experimental procedures were reviewed and approved by the appropriate Animal Use Committee of the Institutions (UCLA and Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, USA, and the University of Pavia, Pavia, Italy) where the experiments were performed. The number of animals used was kept to the minimum necessary for a meaningful interpretation of the data and animal discomfort was kept to the minimum. Male albino, Harlan Porcellus guinea-pigs (Hartley; 250 –350 g; Harlan Laboratories, San Diego, CA, USA; G. Bettinardi; Momo, Novara, Italy) were used.
Electrically induced contractile responses Segments of the distal ileum were dissected from guinea-pig killed by CO2 inhalation and then immersed in standard Tyrode solution composed of (mM): NaCl, 136.9; KCl 2.7; CaCl2 1.8; MgCl2 1.04; NaH2PO4 0.4; NaHCO3 11.9; glucose 5.5. Five animals were used for each experimental condition. LMMP preparations were prepared by teasing the longitudinal muscle with the intact myenteric plexus from the underlying circular muscle (Tonini et al., 1998; McConalogue et al., 1999). For the analysis of electrically induced neurogenic contractions (twitch contractions), 4 cm long strips were folded in half and mounted in organ baths containing 5 ml of Tyrode solution (bubbled with 95% O2 and 5% CO2) at 37 °C, under tension of 5 mN. Isometric contractions were recorded with a force-displacement transducer (Harvard Instruments, South Natick, MA, USA). Following a 45 min equilibration period, strips were stimulated via two platinum electrodes with a Grass S44 stimulator (Grass Instruments Co., Quincy, MA, USA). Square wave pulses (0.5 ms) of supramaximal amplitude (60 V) were delivered at a frequency of 0.1 Hz. Concentration-response curves for the inhibitory effects of porcine galanin (Peninsula, Belmont, CA, USA) (1 nM–10 M), which has similar affinity for each GAL-R subtype, were calculated. The electrically evoked contractions in LMMP preparations were abolished by both tetrodotoxin (1 M) and atropine (1 M), indicating that they are mediated by activation of cholinergic nerves. Non-cumulative concentration-response curves were obtained by half-logarithmic dosing increments. The inhibitory effect of galanin on the stimulated muscle twitch was expressed as percentage of control contractions. The inhibitory effects of galanin were also determined following high (1 M) desensitizing concentrations of galanin (30 min of contact) or in the presence of the GAL-R1 non-peptide selective antagonist, RWJ-57408 (1 M prior to the application of galanin). The binding affinity IC50’s of RWJ-57408 for GAL-R1 ranged between 190 and 2700 nM in human Bowes melanoma cells (Scott et al., 2000). In addition, RWJ-57408 was effective in blocking the galanin inhibition of acetylcholine release in the CNS and the electrically induced, acetylcholine-mediated contraction in the guinea-pig ileum (Scott et al., 2000). Signals were recorded using a PowerLab data acquisition system (Analog Digital Instruments, UK) and analyzed using PowerLab Chart v4.1.1 software.
Data analysis All data are expressed as mean⫾S.E.M. Galanin inhibition curve was analyzed by fitting it to a logistic equation of the form: Effect⫽Emax/1⫹e{⫺2.303⫻slope⫻(log[A]⫺log[A50])} where, Emax⫽ maximum response; [A]⫽molar agonist concentration; [A50]⫽ molar agonist concentration inducing 50% of the maximum response. All data were fitted either to a single logistic expression or to the sum of two logistics. Goodness of fit to a single or double logistic expression was evaluated by the F test of the residual variances using a significance criterion of P⬍0.05 (SAS/STAT User’s guide, release 6.03; SAS Institute Inc., Cary, NC, USA; Lucchelli et al., 1995). Apparent affinity estimate (pKB) from a single antagonist concentration was calculated by the Gaddum (1957) equation.
Induction of peristalsis Peristalsis was studied using a previously described method (Tonini et al., 1981, 1989). Segments of ileum, approximately 7– 8 cm in length, were mounted horizontally in a 50 ml organ bath containing Tyrode solution maintained at 37 °C and gassed with 95% O2 and 5% CO2. Longitudinal movements were detected by means of an isotonic transducer loaded with 1 g. The peristalsis was elicited by delivering Tyrode solution into the lumen, by means of a peristaltic pump, at a constant rate of 0.75 ml/min. The flow was interrupted as soon as peristalsis developed, and the ejected fluid acting by gravity on a balanced switch turned off the
C. Sternini et al. / Neuroscience 125 (2004) 103–112 pump. The arrest of the inflow, controlled by a timer, was maintained for 10 s in order to release pressure on the contracted organ during propulsion, and to allow a period of rest between two peristalses. Intraluminal pressure was measured by two appropriately calibrated pressure transducers (Bell & Howell) connected to the anal end of the preparation, as an indication of circular muscle activity. One pressure transducer was calibrated at high sensitivity to record the threshold pressure for peristalsis and the other at low sensitivity to record the peak pressure (maximal ejection pressure) when the intestine was emptying. Peristaltic activity was studied in 10 min duration cycles; each period of peristaltic activity was followed by at least 15 min of rest.
105
Table 1. Characteristics of primary antibodies Antigen
Host
Dilution
Code and reference
ChAT
Goat
1:100
GAL-R1
Rabbit
1:2000–1:3000
NOS
Mouse
1:100
VIP
Mouse
1:1500
AB144P; Chemicon; Mawe et al., 1996; Li and Furness, 1998 GAL-R1Y225–238; Pham et al., 2002 N31020; Transduction Lab, Lexington, KY, USA; Williamson et al., 1996 VIP55; Wong et al., 1996
Parameters measured The following parameters were examined: a) shortening of the longitudinal muscle during the preparatory phase of peristalsis; b) peak pressure (i.e. maximum ejection pressure) that reflects the anally propagated wave of contraction of the circular muscle responsible for the clearance of the intraluminal content; c) threshold pressure (Pt); d) residual pressure (Pr); e) threshold volume (Vt); f) residual volume (Vr); g) compliance. The threshold pressure was the pressure at the beginning of peristalsis, whereas the residual pressure corresponded to the intraluminal pressure after completion of peristalsis. The threshold volume required to trigger the emptying phase was measured as the volume infused into the intestine plus the residual volume remaining at the end of the previous peristaltic wave. The residual volume, which remained in the intestine after a peristaltic wave, was estimated by deducting the expelled volume from that infused into the lumen during the preparatory (filling) phase. The compliance of the intestinal wall during the preparatory phase (⌬V/⌬P) was defined as the change in intraluminal pressure in response to a given change in intraluminal volume. This value reflects the resistance of the intestinal wall to infused fluid. It can be calculated from the inverse of the slope of the preparatory phase of the intraluminal pressure trace (Waterman et al., 1992) as it follows: average compliance⫽Vt⫺Vr/ Pt⫺Pr⫽⌬V/⌬P, where Vt⫺Vr corresponds to the volume infused during the preparatory phase. Four sets of experiments were carried out. In one set of experiments, the effects of 3 M galanin administered 5 min following the start of the cycle was evaluated. In two other sets, the effect of 1 M or 10 M RWJ-57408, administered at the start of the resting period between cycles, was assessed with and without galanin. In the fourth set of experiments, the effect of 10 M RWJ-57408 was determined in conditions of peristalsis impairment caused by 10 M hexamethonium that blocks the ganglionic nicotinic transmission (Bartho et al., 1989). Signals were recorded as described above.
Data analysis Data are expressed as mean⫾S.E.M. and represent the peristaltic activity recorded for 10 min. The effect of galanin was measured for 5 min following administration. One-way analysis of variance for repeated measures was used for statistical analysis. P⬍0.05 was considered as significant.
Immunohistochemistry Guinea-pigs were deeply anesthetized with i.p. sodium pentobarbital (Nembutal; 100 mg/kg; Abbott Laboratories, Chicago, IL, USA). The distal ileum was dissected, opened up along the mesentery, pinned flat on wax and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4 for 2 h at room temperature and then stored in PB with 0.1% g sodium azide at 4 °C before processing for immunohistochemistry as whole mounts (Sternini et al., 2000). For colchicine treatment, a segment of the distal ileum, pinned flat as described above, was placed in sterile Krebs solution (in mM: KCl, 5.9; NaCl, 118; CaCl2, 2.5; MgSO4, 1.2;
NaH2PO4, 1.4; glucose, 5; Fungizone, 2.5 g/ml; penicillin, 100 IU/ml; and streptomycin, 100 g/ml), bubbled with 95% O2:5% CO2, pH 7.4, at 37 °C, washed and then incubated in Dulbecco’s modified Eagle’s medium (DME/F12) containing 10% v/v fetal bovine serum, penicillin, streptomycin and Fungizone at 37 °C with 95% O2:5% CO2, pH 7.4, for 30 min, transferred to medium containing colchicine (100 M; Sigma Chemical, St. Louis, MO, USA) for 16 –20 h at 37 °C, and then fixed as described above. Tissues were processed by the immunofluorescence method for single and double labeling (Sternini et al., 2000; Pham et al., 2002). Briefly, tissues were washed in PB, incubated in 10% normal donkey serum for 1 h at room temperature to minimize the background, incubated in GAL-R1 antiserum (1:2000 –1:3000) for 3 days at 4 °C, followed by a 2-h incubation in affinity purified donkey anti-rabbit IgG conjugated with fluorescein isothiocyanate (FITC; Jackson Immunolabs, West Grove, PA, USA) at 1:100 dilution, washed and mounted on slides and then coverslipped. The GAL-R1 antiserum has been characterized by immunostaining and Western blot of membrane from cells transfected with GAL-R1 cDNA and from control cells (Pham et al., 2002). For double-label immunofluorescence, tissues were incubated in a mixture of rabbit GAL-R1 antiserum and mouse monoclonal antibody to vasoactive intestinal polypeptide (VIP) at 1:1500 dilution or nitric oxide synthase (NOS; Chemicon, Temecula, CA, USA) at 1:50 dilution; or goat acetylcholine transferase (ChAT; Chemicon) at 1:100 dilution for 3 days at 4 °C, followed by a 2-h incubation at room temperature with a mixture of affinity-purified donkey anti-rabbit IgG coupled with FITC (1:100; Jackson Immunolabs) and donkey antimouse or anti-goat IgG conjugated with Rhodamine Red-X (Red-X; 1:300 –1:500; Jackson Immunolabs), then coverslipped with 90% glycerol containing 2% potassium iodide in 0.1 M PB. Primary and secondary antibodies were diluted in 0.5% Triton X-100 in 0.1 M PB. Specificity controls included preadsorption of the GAL-R1 with the peptide used for immunization (10 M) for 16 –24 h at 4 °C. ChAT, VIP and NOS antibodies (Table 1) have been previously validated for immunohistochemical use in gut tissue (Mawe et al., 1996; Williamson et al., 1996; Wong et al., 1996; Li and Furness, 1998). Three colchicine-treated guinea-pigs were used for cell counting. For each animal, 10 plexuses were randomly selected, the number of myenteric neurons expressing GAL-R1, ChAT, VIP or NOS were counted, and the mean percentage of GAL-R1 neurons containing each of the other marker was calculated. Tissues were examined using a light Leitz Orthoplan Microscope using a Phloem illumination system with “L2” and “N2” filters for FITC and Red-X visualization, and a dualband filter for the simultaneous visualization of FITC and Red-X. A Zeiss laser scanning microscope 410 with a krypton/argon laser and attached to a Zeiss Axiovert 100 microscope with PlanApo 63⫻1.3 na objective was also used. Typically, 10 –20 optical sections were taken with a z-axis of 0.75 m. Images were processed and labeled
106
C. Sternini et al. / Neuroscience 125 (2004) 103–112
RWJ-57408 (1 M), caused a parallel shift to the right of the low potency phase of the galanin curve, without affecting the high potency phase (Fig. 3B). The apparent affinity estimate (pKB) of RWJ-57408 was 6.6. Effects of galanin on peristalsis
Fig. 1. Inhibitory effect of galanin on electrically stimulated twitch contractions of LMMP preparations. Galanin inhibition curve was biphasic. The potency (pEC50(H)) of the first phase was 8.2, while that of the second phase (pEC50(L)) was 6.6. The high potency phase (in the nanomolar concentration range) reached a maximum of about 25%. Each point represents the mean⫾S.E.M. (n⫽5).
using Adobe Photoshop 6.0 (Adobe Systems, Mountain View, CA, USA).
RESULTS Effects of galanin on twitch contractions in LMMP preparations Galanin (1 nM–10 M) induced a concentration-dependent inhibition of electrically stimulated muscle twitch contractions (Fig. 1) up the their abolition. The curve to galanin was better fitted to a biphasic than to a monophasic model (F⫽45.63; P⬍0.001). The potency estimate of the first phase (high potency phase: pEC50(H)) was 8.2, while that of the second low potency phase (pEC50(L)) was 6.6. In unstimulated LMMP preparations, submaximal contraction to acetylcholine (0.1 M) was not affected by 1 M galanin (control: 10.5⫾1.0 mN vs. 9.7⫾1.3 mN in the presence of galanin, n⫽5; Fig. 2). As shown in the tracing, galanin caused a slight contractile response (about 1.5 mN in amplitude). This response and the contractile response produced by acetylcholine were not modified by 1 M tetrodotoxin, a blocker of voltage-dependent neuronal Na⫹-channels (n⫽3), indicating a direct effect on the effector cells, with the galanin contraction probably being mediated by a postjunctional excitatory receptor associated with intracellular calcium mobilization. To desensitize GAL-R, a high (1 M) concentration of galanin was used and left in the bath for 30 min. The initial inhibitory effect of galanin on twitch contractions (approximately 70%) gradually vanished with a recovery of control twitch amplitude by 90% indicating receptor desensitization. Under these conditions, the high potency phase of the galanin inhibitory curve was completely suppressed and the curve became monophasic. The low potency phase was also reduced without suppression of the maximum response (Fig. 3A). By contrast, the GAL-R1 antagonist,
Intraluminal fluid infusion produced the two phases of peristalsis, a preparatory phase followed by an emptying phase. During the preparatory phase, there was a slight and slow increase of the intraluminal pressure due to the accommodation processes of the circular muscle (Waterman et al., 1994a). At the same time, there was a shortening of the longitudinal muscle, which reached the maximum when the emptying phase occurred. This phase was characterized by a threshold pressure of approximately 100 Pa, a threshold volume of about 900 l, and a peak pressure (maximum ejection pressure) of about 1800 Pa. The frequency of peristaltic waves in a 10 min-cycle was 8.4⫾0.4 (n⫽11; Fig. 4). Galanin (3 M) significantly reduced the longitudinal muscle contraction (P⬍0.05) during the preparatory phase of peristalsis (Fig. 5) and the peak pressure (P⬍0.002; Fig. 6) generated during the emptying phase. The inhibitory effect of galanin on longitudinal muscle contractions and peak pressure was antagonized by RWJ-57408 (1 and 10 M). RWJ-57408 per se did not have any effect on these nor on the following parameters (not shown). Galanin (3 M) did not affect the threshold volume (Fig. 7) and pressure (Fig. 8) required to trigger the emptying phase of peristalsis, but significantly increased the residual volume (P⬍0.002) and pressure (P⬍0.05) after completion of a peristaltic event. RWJ-57408 antagonized these galanin effects at 10 M, but not at 1 M (Figs. 7, 8). Galanin (3 M) decreased the compliance of the intestinal wall by 25%. This effect was antagonized by the GAL-R1 antagonist at 10 M (P⬍0.06 vs. control; Fig. 9). The lack of effect of RWJ-57408 per se on peristalsis casts doubt on the
Fig. 2. Effect of galanin on exogenously applied acetylcholine (ACh: 0.1 M) in resting LMMP preparations. Note that a concentration of galanin (1 M) that induces ⬎50% inhibition of twitch contractions does not affect the contractile response induced by exogenous acetylcholine. Dots indicate drug administration.
C. Sternini et al. / Neuroscience 125 (2004) 103–112
107
Fig. 3. (A) Inhibitory effect of galanin on electrically stimulated contractions of LMMP preparations (open circle) or following self-desensitization with a high concentration (1 M for 30 min) of galanin (closed triangle). Galanin inhibited the amplitude of twitch contractions up to their abolition. Note that the desensitizing concentration of galanin completely suppressed the high potency phase of galanin curve, and partially reduced the low potency phase without abolishing its maximum. (B) Galanin inhibitory curve in the presence (closed circle) and absence (open circle) of the GAL-R1 antagonist (RWJ-57408: 1 M). Note that the GAL-R1 antagonist inhibits the low potency phase of galanin curve, whereas it did not affect the high potency phase. Values are expressed as mean⫾S.E.M. (n⫽5).
functional role of endogenous galanin in modulating propulsive activity. On the other hand, a subtle tuning of
neuronal activity cannot be detected with an intact cholinergic transmission, which is the main excitatory trans-
Fig. 4. Recording of peristalsis in the guinea-pig isolated ileum during a 10 min cycle. The top tracing represents the shortening (mm) of the longitudinal muscle, whereas the middle and bottom tracings illustrate the intraluminal pressure (Pascal: Pa) recorded at high and low sensitivity, respectively. Arrows indicate the threshold (Pt) and residual (Pr) pressure (middle tracing) and the peak pressure (bottom tracing). The peak pressure in the middle tracing is truncated because it is off scale.
108
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Fig. 5. Effect of galanin (Gal: 3 M) on longitudinal muscle shortening (mm) during the preparatory (filling) phase of peristalsis in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates longitudinal muscle shortening in the absence of drug treatment. Galanin significantly inhibits the shortening of longitudinal muscle through a mechanism antagonized by RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
mission regulating peristalsis. To overcome this problem, a series of experiments were conducted in the presence of the ganglionic blocking agent, hexamethonium. Hexamethonium induced an increase of the threshold and residual volume at the end of the preparatory phase of peristalsis (Fig. 10), which are considered a representative measure of peristalsis impairment (Waterman et al., 1992). Even under these conditions of partial blockade of cholinergic transmission, which should facilitate the detection of small modulatory effects of endogenous galanin on GAL-R1, RWJ-57408 at 10 M did not have any effect (Fig. 10).
Fig. 6. Effect of galanin (Gal: 3 M) on peak pressure (Pascal: Pa) of each peristaltic wave in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates pressure values in the absence of drug treatment. Galanin significantly inhibits the peak pressure through a mechanism antagonized by RWJ-57408 (10 M). Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control. Note that RWJ-57408 treatment (10 M) in the presence of galanin causes a response, which is not different from control response indicating a full antagonism of galanin inhibitory effect.
Fig. 7. Effect of galanin (Gal: 3 M) on the threshold (Vt) and residual (Vr) volume (l) in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates volume values in the absence of drug treatment. Galanin did not affect the threshold volume, but significantly increased the residual volume through a mechanism antagonized by 10 M RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
GAL-R1 immunoreactivity in the myenteric plexus of the guinea-pig ileum GAL-R1 immunoreactivity was observed in numerous neurons of the myenteric plexus (Fig. 11), and in fibers within the plexuses and distributed to the muscle layer (not shown) with a pattern similar to the one described in the rat gastrointestinal tract (Pham et al., 2002). Confocal microscopy was required to discern cell bodies due to the dense neuropil formed by GAL-R1 immunoreactive fibers in the myenteric plexus. Double labeling studies revealed that the vast majority (about 80%) of GAL-R1 myenteric neurons contained ChAT immunoreactivity (Fig. 11). However, there were also GAL-R1 myenteric neurons immunoreactive for VIP or NOS (about 12% and 18%, respectively; Fig. 11).
Fig. 8. Effect of galanin (Gal: 3 M) on the threshold (Pt) and residual (Pr) pressure (Pascal: Pa) in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates pressure values in the absence of drug treatment. Galanin did not affect the threshold pressure, but significantly increased the residual pressure through a mechanism antagonized by 10 M RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Fig. 9. Effect of galanin (Gal: 3 M) on circular muscle compliance (L/Pa) in the absence (n⫽8) and in the presence of the GAL-R1 antagonist, RWJ-57408 (RWJ) 1 M (n⫽5) or 10 M (n⫽6). Control (C, n⫽11) indicates compliance values in the absence of drug treatment. Galanin significantly reduces the compliance through a mechanism antagonized by 10 M RWJ-57408. Data are expressed as mean⫾S.E.M. * P⬍0.05 vs. control.
DISCUSSION These studies demonstrate that 1) galanin inhibits cholinergic transmission to the longitudinal muscle by acting at a prejunctional level via the activation of at least two receptors, one being GAL-R1, since the low potency phase of the galanin inhibitory curve is antagonized by the GAL-R1 antagonist, RWJ-57408; 2) galanin affects peristalsis by reducing its efficiency (reduction of longitudinal muscle contraction and peak pressure) and decreasing the compliance of the intestinal wall through activation of GAL-R1; and 3) GAL-R1 immunoreactivity is localized to functionally distinct neurons of both the ascending and descending pathways. Galanin effect on neurogenic twitch contractions was prejunctional, since submaximal contractions to exogenous acetylcholine in unstimulated LMMP preparations were not
Fig. 10. Effect of hexamethonium (Hex: 10 M) on threshold and residual volume. Control C indicates volume in the absence of any treatment. Note that the significant increase of threshold and residual volume induced by hexamethonium is not affected by RWJ-57408 (RWJ) at 10 M (n⫽8). * P⬍0.05 vs. control.
109
affected. The inhibitory curve caused by galanin was biphasic, with the low potency phase antagonized by RWJ-57408 and the high potency phase abolished by GAL-R desensitization. This is consistent with the involvement of two receptors/mechanisms including the GAL-R1. It is likely that the high potency phase (approximately 25% of the total curve) is mediated by the GAL-R3, an inhibitory receptor, which utilizes the same transduction mechanisms as the GAL-R1 (Smith et al., 1998), although selective antagonists at the GAL-R3 are required to clarify this point. Indeed, GAL-R1 and GAL-R3 mediate pertussis toxin sensitive inhibition of adenylate cyclase. The other receptor activated by galanin, the GAL-R2, is an excitatory receptor mainly coupled to Gq/11 signaling pathway (Smith et al., 1997; Wang et al., 1998a). This receptor induces direct muscle contraction and transmitter release; therefore it cannot be responsible for the inhibitory effect caused by nanomolar concentrations of galanin. It has been reported that galanin inhibits electrically induced neurogenic cholinergic contractions and acetylcholine release in the guinea-pig ileum (Ekblad et al., 1985a; Yau et al., 1986; Mulholland et al., 1992; Akehira et al., 1995). Our findings extend previous results, indicating that the inhibitory effect of galanin is mediated by two distinct receptors, which undergo different degree of desensitization. The morphological data showing that the vast majority of GAL-R1 neurons are cholinergic support the conclusion that this receptor plays a major role in the galanin inhibition of excitatory neuromuscular transmission. Peristalsis is a polarized motor event that propels the intestinal contents anally and depends entirely on enteric neurons (Kosterlitz and Lees, 1964; Tonini et al., 1981; Waterman et al., 1992). Distension of the gut or mucosal stimulation induces oral reflex contraction and aboral relaxation. These reflexes are initiated by sensory neurons, which make contact with ascending and descending neurons (Waterman et al., 1994b). In the isolated small intestine, peristalsis consists of a preparatory phase during which intraluminal fluid infusion provides a radial mechanical stimulus to the circular muscle up to a threshold volume that triggers the emptying phase. This phase initiates at the most oral end of the preparation with an anally propagating circular muscle contraction causing fluid displacement (Tonini et al., 1981; Waterman et al., 1992, 1994b). Galanin reduces peristaltic efficiency in the guinea-pig ileum by affecting several parameters of peristalsis. It inhibits the longitudinal muscle shortening and the peak pressure through a pathway involving GAL-R1 as indicated by the RWJ-57408 antagonism. Since shortening of longitudinal muscle and peak pressure are mediated by excitatory neurons, the observations that they are inhibited by RWJ-57408 are in agreement with our findings with LMMPs supporting a major role of GAL-R1 in the galanin inhibition of excitatory neuromuscular transmission. In addition to a reduction of the longitudinal muscle contraction and peak pressure, galanin induced an increase of the residual volume and pressure, which indicates that part of the intraluminal content of the intestine cannot be expelled. Both these parameters are antagonized by 10 M RWJ-
110
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Fig. 11. Confocal images of myenteric plexuses of the guinea-pig ileum showing GAL-R1 at the cell surface (green fluorescence), ChAT, VIP and NOS (red fluorescence) immunoreactivity in the cytoplasm of myenteric neurons. GAL-R1&ChAT, GAL-R1&VIP, and GAL-R1&NOS: simultaneous visualization of GAL-R1 and ChAT, VIP or NOS immunoreactivity. Arrows point to examples of neurons that contain GAL-R1 and ChAT, VIP or NOS immunoreactivity. Scale bars: GAL-R1&ChAT⫽10 m; GAL-R1&VIP⫽20 m; GAL-R1&NOS⫽15 m.
57408. This suggests that GAL-R1 participates in the mechanisms responsible for the intraluminal clearance. Galanin decreases the compliance of the circular muscle during the preparatory phase of peristalsis. Compliance reflects the resistance of the intestinal wall to the infused intraluminal fluid and it is controlled by nonadrenergic, noncholinergic inhibitory (nitrergic) pathways (Waterman et al., 1994a). The inhibitory effect of galanin on compliance was affected by RWJ-57408, suggesting the involvement of GAL-R1. This is in agreement with the presence of GAL-R1 on descending myenteric neurons, as shown by the colocalization of GAL-R1 with VIP and NOS immunoreactivity. However, we cannot exclude the participation of an excitatory receptor (like the GAL-R2) located on smooth muscle cells, which may partly hinder the accommodation of the circular muscle during the preparatory phase of peristalsis. The failure of RWJ-57408 to affect peristalsis per se suggests that endogenous galanin does not exert a functional tonic inhibition on neuronal pathways subserving
peristalsis through GAL-R1. However, the inhibitory action of galanin could be undetectable when excitatory cholinergic transmission is intact. The lack of effect of RWJ57408 on peristalsis when cholinergic nicotinic transmission is partially blocked by hexamethonium excludes a modulatory role of endogenous galanin on neuronal pathways controlling peristalsis. This implies that in our conditions, galanin exerts a pharmacological rather than a physiological role on peristaltic activity. GAL-R1 is localized to functionally distinct populations of enteric neurons and to fibers distributed to the muscle. This pattern is comparable to the one we have described for the rat stomach and small intestine (Pham et al., 2002). The distribution of GAL-R1 neurons and their sites of innervation are in accord with the galanin distribution (Ekblad et al., 1985b; Burgevin et al., 1995; Pham et al., 2002). The presence of GAL-R1 on neuronal cell surface and processes distributed to the muscle is consistent with pre- and post-synaptic actions of galanin via this receptor and supports the concept that GAL-R1 is transported to
C. Sternini et al. / Neuroscience 125 (2004) 103–112
axon terminals. In the enteric nervous system, neurons can be differentiated on the basis of their morphology and projections, and of the combination of transmitters that they synthesize and release (Costa et al., 1996). The finding that as many as 80% of GAL-R1 myenteric neurons contain ChAT supports the concept that GAL-R1 bearing myenteric neurons comprise ascending excitatory neurons. This is in agreement with the functional data showing that galanin inhibits the contractile response to electrical stimulation (this paper; Ekblad et al., 1985a; Yau et al., 1986; Mulholland et al., 1992; Akehira et al., 1995) through the inhibition of excitatory transmitters like acetylcholine and substance P via GAL-R1. The observation that the inhibitory effect of galanin on stimulated acetylcholine release is reversed by pertussis toxin further supports the involvement of GAL-R1. However, the presence of GAL-R1 on VIP and NOS immunoreactive neurons, in addition to ChAT containing neurons, is consistent with a modulatory role of this receptor on both the ascending and descending pathways. In summary, on the basis of our functional and morphological studies, we can conclude that galanin effects on neuromuscular transmission in the longitudinal muscle and on peristaltic activity in the guinea-pig ileum involve the activation of GAL-R1 on ascending and descending pathways. The failure of RWJ-57408 to modify per se cholinergic neuromuscular transmission and peristalsis, despite a redundant presence of GAL-R1 in myenteric neurons, suggests that under our experimental conditions endogenous galanin does not exert a tonically active modulatory role on gut motility via GAL-R1. Acknowledgements—This paper was supported by National Institutes of Health grants, DKs 57037 (to C.S.), 35740 (to C.S.), 41301 (CURE:DDRCC: Morphology/Imaging Core to C.S.), and University of Pavia, Italy Funds (to M.T.).
REFERENCES Akehira K, Nakane Y, Hioki K, Taniyama K (1995) Site of action of galanin in the cholinergic transmission of guinea pig small intestine. Eur J Pharmacol 284:149 –155. Bartfai T, Fisone G, Langel U (1992) Galanin and galanin antagonists: molecular and biochemical perspectives. Trends Pharmacol Sci 13:312–317. Bartho L, Holzer P, Leander S, Lembeck F (1989) Evidence for an involvement of substance P, but not cholecystokinin-like peptides, in hexamethonium-resistant intestinal peristalsis. Neuroscience 28: 211–217. Bauer FE, Zintel A, Kenny MJ, Calder D, Ghatei MA, Bloom SR (1989) Inhibitory effect of galanin on postprandial gastrointestinal motility and gut hormone release in humans. Gastroenterology 97:260 – 264. Bedecs K, Berthold M, Bartfai T (1995) Galanin-10 years with a neuroendocrine peptide. Int J Biochem Cell Biol 27:337–349. Branchek T, Smith KE, Walker MW (1998) Molecular biology and pharmacology of galanin receptors. Ann NY Acad Sci 863:94 –107. Branchek TA, Smith KE, Gerald C, Walker MW (2000) Galanin receptor subtypes. Trends Pharmacol Sci 21:109 –117. Burgevin MC, Loquet I, Quarteronet D, Habert-Ortoli E (1995) Cloning, pharmacological characterization, and anatomical distribution of a rat cDNA encoding for a galanin receptor. J Mol Neurosci 6:33–41. Costa M, Brookes SJH, Steele PA, Gibbins I, Burcher E, Kandiah CJ
111
(1996) Neurochemical classification of myenteric neurons in the guinea pig ileum. Neuroscience 75:949 –967. Ekblad E, Hakanson R, Sundler F, Walhestedt C (1985a) Galanin: neuromodulatory and direct contractile effects on smooth muscle preparations. Br J Pharmacol 86:241–246. Ekblad E, Rokaeus A, Hakanson R, Sundler F (1985b) Galanin nerve fibers in the rat gut: distribution, origin and projections. Neuroscience 16:355–363. Floren A, Land T, Langel U (2000) Galanin receptor subtypes and ligand binding. Neuropeptides 34:331–337. Fox JET, McDonald TJ, Kostolanska F, Tatemotot K (1986) Galanin: an inhibitory neural peptide of the canine small intestine. Life Sci 39:103–110. Furness JB, Costa M, Rokaeus A, McDonald TJ, Brooks B (1987) Galanin-immunoreactive neurons in the guinea-pig small intestine: their projections and relationships to other enteric neurons. Cell Tissue Res 250:607–615. Gaddum JH (1957) Series of drug antagonism. Pharmacol Rev 9:211– 218. Gu ZF, Pradhan TK, Coy DH, Jensen RT (1995) Interaction of galanin fragments with galanin receptors on isolated smooth muscle cells from guinea pig stomach: identification of a novel galanin receptor subtype. J Pharmacol Exp Ther 272:371–378. Kosterlitz HW, Lees GM (1964) Pharmacological analysis of intrinsic intestinal reflexes. Pharmacol Rev 16:301–339. Li ZS, Furness JH (1998) The immunohistochemical localization of cholinergic markers in putative intrinsic primary afferent neurons of the guinea pig small intestine. Cell Tissue Res 294:35–43. Lucchelli A, Santagostino-Barbone MG, Barbieri A, Candura SM, Tonini M (1995) The interaction of antidepressant drugs with central and peripheral (enteric) 5-HT3 and 5-HT4 receptors. Br J Pharmacol 114:117–125. Mawe GM, Talmage EK, Lee KP, Parsons RL (1996) Expression of choline acetyltransferase immunoreactivity in guinea pig cardiac ganglia. Cell Tissue Res 285:281–286. McConalogue K, Grady EF, Minnis J, Balestra B, Tonini M, Brecha NC, Bunnett NW, Sternini C (1999) Activation and internalization of the opioid receptor by the newly described endogenous agonists, endomorphin-1 and endomorphin-2. Neuroscience 90:1051–1059. Melander T, Hokfelt T, Rokaeus A, Fahrenkrug J, Tatemoto K, Mutt V (1985) Distribution of galanin-like immunoreactivity in the gastrointestinal tract of several mammalian species. Cell Tissue Res 239:253–270. Mulholland MW, Schoeneich S, Flowe K (1992) Galanin inhibition of enteric cholinergic neurotransmission: guanosine triphosphatebinding protein interactions with adenylate cyclase. Surgery 112: 195–201. Muramatsu I, Yanaihara N (1988) Contribution of galanin to noncholinergic, non-adrenergic transmission in rat ileum. Br J Pharmacol 94:1241–1249. Palmer JM, Schemmann M, Tamura K, Wood JD (1986) Galanin mimics slow synaptic inhibition in myenteric neurons. Eur J Pharmacol 124:379 –380. Pham T, Guerrini S, Wong E, Reeve JR Jr, Sternini C (2002) Distribution of galanin receptor 1 immunoreactivity in the rat stomach and small intestine. J Comp Neurol 450:292–302. Rattan S (1991) Role of galanin in the gut. Gastroenterology 100: 1762–1768. Scott MK, Ross TM, Lee DH, Wang HY, Shank RP, Wild KD, Davis CB, Crooke JJ, Potocki AC, Reitz AB (2000) 2,3-Dihydro-dithiin and -dithiepine-1,1,4,4-tetroxides: small molecule non-peptide antagonists of the human galanin hGAL-1 receptor. Bioorg Med Chem 8:1383–1391. Smith KE, Forray C, Walker MW, Jones KA, Tamm JA, Bard J, Branchek TA, Linemeyer DL, Gerald C (1997) Expression cloning of a rat hypothalamic galanin receptor coupled to phosphoinositide turnover. J Biol Chem 272:24612–24616. Smith KE, Walker MW, Artymyshyn R, Bard J, Borowsky B, Tamm JA,
112
C. Sternini et al. / Neuroscience 125 (2004) 103–112
Yao W-J, Vaysse PJ-J, Branchek TA, Gerald C, Jones KA (1998) Galanin GALR3 receptors: cloning and functional expression of rat and human receptors. J Biol Chem 273:233321–233326. Sternini C, Brecha NC, Minnis J, D’Agostino G, Balestra B, Fiori E, Tonini M (2000) Role of agonist-dependent receptor internalization in the regulation of opioid receptors. Neuroscience 98:233–241. Tamura K, Palmer JM, Wood JD (1987) Galanin suppresses nicotinic synaptic transmission in the myenteric plexus of the guinea pig ileum. Eur J Parmacol 136:445–446. Tonini M, Fiori E, Balestra B, Spelta V, D’Agostino G, Di Nucci A, Brecha N, Sternini C (1998) Endomorphin-1 and endomorphin-2 activate opioid receptors in myenteric neurons of the guinea-pig small intestine. Naunyn Schmiedebergs Arch Pharmacol 358:686–689. Tonini M, Frigo G, Lecchini S, D’Angelo L, Crema A (1981) Hyoscineresistant peristalsis in guinea-pig ileum. Eur J Parmacol 71:375–381. Tonini M, Galligan JJ, North RA (1989) Effects of cisapride on cholinergic neurotransmission and propulsive motility in the guinea pig ileum. Gastroenterology 96:1257–1264. Wang S, Hashemi T, Fried S, Clemmons AL, Hawes BE (1998a) Differential intracellular signaling of the GalR1 and GalR2 galanin receptor subtypes. Biochemistry 37:6711–6717. Wang YF, Mao YK, Fox-Threlkeld JE, McDonald TJ, Daniel EE (1998b) Colocalization of inhibitory mediators, NO, VIP and galanin, in canine enteric nerves. Peptides 19:99 –112.
Waterman SA, Costa M, Tonini M (1992) Modulation of peristalsis in the guinea-pig isolated small intestine by exogenous and endogenous opioids. Br J Pharmacol 106:1004 –1010. Waterman SA, Costa M, Tonini M (1994a) Accommodation mediated by enteric inhibitory reflexes in the isolated guinea-pig small intestine. J Physiol (Lond) 474:539 –546. Waterman SA, Tonini M, Costa M (1994b) The role of ascending excitatory and descending inhibitory pathways in peristalsis in the isolated guinea-pig small intestine. J Physiol (Lond) 481:223–232. Waters SM, Krause JE (2000) Distribution of galanin-1, -2 and -3 receptor messenger RNAs in central and peripheral tissues. Neuroscience 95:265–271. Williamson S, Pompolo S, Furness JB (1996) GABA and nitric oxide synthase immunoreactivities are colocalized in a subset of inhibitory motor neurons of the guinea pig small intestine. Cell Tissue Res 284:29 –37. Wong HC, Sternini C, Lloyd K, De Giorgio R, Walsh JH (1996) Monoclonal antibody to VIP: production, characterization, immunoneutralizing activity, and usefulness in cytochemical staining. Hybridoma 15:133–139. Yau WM, Dorsett JA, Youther ML (1986) Evidence for galanin as an inhibitory neuropeptide on myenteric cholinergic neurons in the guinea pig small intestine. Neurosci Lett 72:305–308.
(Accepted 18 December 2003)