relaxant function and mRNA detection in smooth muscle cells isolated ...

6 downloads 0 Views 375KB Size Report
Aug 28, 2004 - dependent manner, both human taenia coli and circular colonic smooth ... les cellules des muscles lisses circulaires et des tænia coli du côlon ...
515

␤3-Adrenoceptors: relaxant function and mRNA detection in smooth muscle cells isolated from the human colon Carola Severi, Ivan Tattoli, Giovanna Romano, Vito D. Corleto, and Gianfranco Delle Fave

Abstract: Functional involvement of β 3 -adrenoceptors in controlling human gastrointestinal motility has not been unequivocally assessed yet. The direct myogenic contribution of these receptors was examined, by in vitro functional studies and analysis of mRNA expression, on smooth muscle cells separately isolated from taenia coli and circular muscle layers of the human colon. Isoproterenol, a nonselective β-adrenoceptor agonist, relaxed, in a concentrationdependent manner, both human taenia coli and circular colonic smooth muscle cells, although displaying a higher intrinsic activity (65.3 ± 2.3 vs. 55.2 ± 1.4% maximal relaxation) and potency (pEC50: 7.41 ± 0.07 vs. 6.32 ± 0.08) were greater on taenia coli than circular cells. In the presence of the β 1 -antagonist CGP20712A and of the β 2 -antagonist ICI 118,551, a 25–30% decrease in isoproterenol intrinsic activity was observed on both cell types and on taenia coli, the nonselective β 1 /β 2 -antagonist propranolol produced a rightward shift of the isoproterenol concentration-response curve with mean estimated pKB values (8.12 ± 0.27 at 0.1 µM and 6.45 ± 0.13 at 1 µM) lower than that expected for both β 1 - and β 2 -adrenoceptors. CGP12177A and SR 58611A, two β 3 -adrenoceptor agonists, presented an intrinsic activity comparable to that of isoproterenol in the presence of β 1 - and β 2 -antagonists, the former being more potent on taenia coli than on circular smooth muscle cells. β 3 -Adrenoceptor mRNA was detected by reverse transcription PCR on both cell types. These results strongly suggest a direct functional role of β 3 -adrenoceptors in the human colon. Key words: adrenoceptors, β 3 -adrenoceptors, smooth muscle cells, taenia coli, human colon. Résumé : L’implication fonctionnelle des récepteurs adrénergiques β 3 dans le contrôle de la motilité gastro-intestinale chez l’humain n’a pas encore été établie sans équivoque. La contribution myogénique directe de ces récepteurs a été examinée par des analyses fonctionnelles in vitro et par des analyses d’expression de l’ARNm chez des cellules des muscles lisses isolées séparément des tænia coli ou des couches musculaires circulaires du côlon humain. L’isoprotérénol, un agoniste non sélectif des β-adrénorécepteurs, a permis de relâcher, d’une manière dose-dépendante, les cellules des muscles lisses circulaires et des tænia coli du côlon humain, bien qu’il ait eu une activité intrinsèque (65,3 ± 2,3 vs 55,2 ± 1,4 % de relâchement maximal) et une puissance (pEC50 7,41 ± 0,07 vs 6,32 ± 0,08) supérieures sur les cellules des tænia coli. En présence de l’agoniste β 1 CGP20712A et de l’agoniste β 2 ICI 118,551, une diminution de 25 à 30 % de l’activité intrinsèque de l’isoprotérénol a été observée chez les deux types cellulaires. Sur les tænia coli, le propranolol, un antagoniste β 1 /β 2 , a déporté la courbe de réponse à l’isoproténérol vers la droite et a résulté en des valeurs moyennes de pKB (8,12 ± 0,27 à 0,1 µM et 6,45 ± 0,13 à 1 µM) plus faibles qu’attendues pour les adrénorécepteurs β 1 et β 2 . Le CGP12177A et le SR 58611A, 2 agonistes des adrénorécepteurs β 3 , ont affiché une activité intrinsèque comparable à celle de l’isoprotérénol en présence des antagonistes b1 et β 2 , le premier étant plus actif sur les tænia coli que sur les cellules des muscles circulaires lisses. L’ARNm de l’adrénorécepteur β 3 a été détecté par RT-PCR chez les 2 types de cellules. Ces résultats suggèrent fortement un rôle fonctionnel direct des adrénorécepteurs β 3 dans le côlon humain. Mots clés : adrénorécepteurs, β 3 adrénorécepteurs, cellules des muscles lisses, tænia coli, côlon humain. [Traduit par la Rédaction]

Severi et al.

522

Introduction The existence of β 3-adrenoceptors is now generally accepted, and these receptors have been shown to be abundant

in a number of gastrointestinal smooth muscle preparations (Summers 1999). In the control of gut motility, the prominent role played by β 3-adrenoceptors has been recognized both in animals (van der Vliet et al. 1990; Landi et al. 1993;

Received 22 December 2003. Published on the NRC Research Press Web site at http://cjpp.nrc.ca on 28 August 2004. C. Severi,1 I. Tattoli, G. Romano, V.D. Corleto, and G. Delle Fave. Department of Clinical Sciences, Clinica Medica 2, Policlinico Umberto I, Viale del Policlinico, “La Sapienza” University of Rome, 00161 Roma, Italy. 1

Corresponding author (e-mail: [email protected]).

Can. J. Physiol. Pharmacol. 82: 515–522 (2004)

doi: 10.1139/Y04-042

© 2004 NRC Canada

516

Manara et al. 1995, 2000; Lezama et al. 1996; Koike et al. 1997) and humans (De Ponti et al. 1996, 1999; Bardou et al.1998; Kelly et al. 1998). Immunohistochemistry (Anthony et al. 1998) and molecular biology (mRNA) (Krief et al. 1993; Berkowitz et al. 1995) studies have confirmed the presence of β 3-adrenoceptors, a β-adrenoceptor subtype, in the smooth muscle of the human alimentary tract. Sympathetic nerves reach all regions of the human gastrointestinal tract (Furness and Costa 1974), inducing relaxation of the nonsphincteric gastrointestinal smooth muscle. Impairment of gut motility due to involvement of the sympathetic nerves is believed to play a role in several alimentary disorders, such as irritable bowel syndrome (Lyrenas 1985), autonomic neuropathy of Crohn’s disease (Lindgren et al. 1991), and postoperative ileus (Hallerback et al. 1987). Understanding the functional involvement of the different adrenoceptors in controlling gastrointestinal motility appears to be essential for the development of new compounds to treat gut dismotility. The inhibitory effect exerted by sympathetic nerves on gut motility is in part neurally mediated, by the reduction of the activity of the myenteric motor neurons (Burnstock and Wong 1981), and in part directly mediated, by the stimulation of membrane adrenoceptors on smooth muscle cells (SMC) (Luckensmeyer and Keast 1998). The direct effect on the nonsphincteric gastrointestinal muscle appears to be predominantly β-adrenoceptor-mediated (β1, β 2, and β 3), with the relative contribution of each subtype varying according to the preparation examined (Luckensmeyer and Keast 1998; Roberts et al. 1999; Oostendorp et al. 2000). In humans, most research has been focused on the large intestine, the region of the gut that contains a high density of sympathetic terminals directly innervating the smooth muscle layers (Norberg 1964). It has been suggested that β1-adrenoceptors are located within the ganglionic plexus, and that β 2adrenoceptors are located in SMC (Ek et al. 1986). Controversial results concerning the involvement of β 3adrenoceptors have been obtained in studies of circular smooth muscle relaxation (Roberts et al. 1997; De Ponti et al. 1999; Manara et al. 2000); however, functional evidence of β 3-adrenoceptor involvement has been reported by De Ponti et al. (1999). Molecular studies have failed to determine the exact expression of β 3-adrenoceptors in circular colonic muscle, and have provided evidence only from a mixed longitudinal/circular muscle coat (Roberts et al. 1997). The aim of this study was to examine the direct effects of β3-adrenoceptors on SMC isolated separately from taenia coli and circular muscle layers of the human colon, and to determine, separately in each cell type, the mRNA coding for the β 3-adrenoceptor subtype by means of reverse transcription – polymerase chain reaction (RT-PCR).

Materials and methods Materials Compounds were obtained from the following sources: collagenase (CLS type II) and soybean trypsin inhibitor from Worthington (Freehold, N.J.); Eagle’s minimum essential amino acid mixture from GIBCO (Paisley, UK); carbamylcholine (carbachol), papaverine hydrochloride, (–)-isoproterenol hydrochloride, (±)-propranolol hydrochloride, (±)-1-

Can. J. Physiol. Pharmacol. Vol. 82, 2004

[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol (ICI 118,551), (±)-2-hydroxy-5-[2-[[2hydroxy-3-[4-[1-methyl-4-(trifluoromethyl)-1H-imidazol-2yl]phenoxy]propyl]amino]ethoxy]-benzamide methanesulfonate (CGP20712A), ascorbic acid, bovine serum albumin (BSA) (fraction V), N-(2-hydroxyethyl ) piparazine-N ′-2-ethanesulfonic acid (HEPES) from Sigma (St. Louis, Mo.); (±)-4-[(3t-butylamino-2-hydroxypropoxy)-benzimidazol-2-one hydrochloride] (GCP12177A), and (N [(25)- 7-carbethoxy-1,2,3,4tetrahydronaphth-2-yl]-(2R)-2-hydroxy-2-(3-chlorophenil) ethanamine hydrochloride) (SR 58611A) from Sanofi– Synthelabo (Research Center, Milan, Italy). Isoproterenol solutions were prepared in 0.1% BSA HEPES medium buffer containing 0.1% ascorbic acid; the remaining drugs were prepared in 0.1% BSA – HEPES medium buffer alone. Tissues preparations and dispersion of smooth muscle cells SMC were separately isolated from the circular and taenia coli muscle layers of the distal colon (sigma) in 12 surgical specimens. Segments (4–6 cm2) of normal colon were obtained from 12 patients (6 females, 6 males; age range, 45– 78 years), who underwent colectomy for colonic carcinoma, and who had not been previously treated with radio- or chemotherapy. All patients gave informed consent. Specimens of whole colon were taken immediately after surgical removal from the unobstructed macroscopically normal area, approximately 7–8 cm from the neoplastic area. Immediately after removal, the segments were placed in iced HEPES buffer bubbled with O2, and the mucosa and submucosa were removed from samples with scissors. Slices of muscle 0.5-mm thick were separately excised from taenia and intertaenial regions with a Stadie-Riggs tissue slicer (Thomas Scientific Apparatus, Philadelphia, Pa.). The intertaenial sample was placed, submucosa-face down, in the slicer, and the bottom slice containing the submucosa and the top two slices containing the serosa and the outer muscle layer were discarded. Slices of the taenia coli muscle (0.5 mm thick) were obtained from the inner part of the taenial samples, using the same tissue slicer. Slices were cut into strips 1 cm in length and 3 mm in width. Cells were isolated separately from each region, using a modified version of a technique described elsewhere (Bitar and Makhlouf 1982a). Muscle strips were incubated at 31 °C for successive 45- and 60-min periods in 15 mL of HEPES medium, containing 0.1% collagenase (150 U/mL). The composition of the medium was as follows (in mmol/L): NaCl, 115; KCl, 5.8; KH2PO4, 2.1; CaCl2, 2.0; MgCl2, 0.6; HEPES, 25; and glucose, 14. The medium also contained 0.01% soybean trypsin inhibitor and 2.1% essential amino acid mixture. At the end of the second incubation period, strips were washed with an enzyme-free medium on 500-µm Nitex mesh, and reincubated in the same medium for 30 min to allow the spontaneous dispersion of cells. In some experiments, cell dispersion was accelerated by gentlye suctioning the suspension with an inverted 10-mL pipette. The cells were then filtered through 500-µm Nitex mesh and harvested. Measurement of response in isolated smooth muscle cells Contraction and relaxation were assessed in a suspension of SMC, as described elsewhere (Bitar and Makhlouf 1982a, © 2004 NRC Canada

Severi et al.

1982b). For contraction studies, 0.5 mL of cell suspension was added to 0.2 mL of the incubation medium, which contained a maximal concentration of carbachol (30 nmol/L). The reaction was interrupted after 30 s with acrolein (final concentration 1%), the interval timed to elicit a peak contraction. Relaxation was measured by preincubating the cell suspension with the relaxant agent for 60 s, adding a maximal concentration of carbachol, and then interrupting the reaction with acrolein after 30 s. To measure control-cell length, the relaxant agent was replaced with an equivalent volume of medium. In the control state, and after the addition of the compounds tested, the length of 50 cells in sequential microscopic fields was measured by imagescanning micrometry (Lasico, Los Angeles, Calif.). Contraction was expressed as the percent decrease in cell length from control (taken as 100); relaxation was expressed as the percent inhibition of maximal contraction induced by the contractile agent.

517

agarose gel β 3- adrenergic receptor products and controls were transferred to a nitrocellulose filter and hybridized to (γ-32P)-labelled β 3-adrenergic receptor oligonucleotide. The nitrocellulose filter was washed at high stringency, air dried, and exposed to X-Omat film (Kodak) for 6–12 h. Data analysis Results are expressed as means ± SEM of duplicate examinations of n experiments. All n values refer to the number of individual patients from whom colon specimens were obtained. Data were statistically evaluated using Student’s t test for paired and unpaired data. p values < 0.05 were considered statistically significant. pEC50 values, corresponding to the negative logarithm to base 10 of the EC50 (the agonist concentration required to produce 50% of the maximal relaxation effect), were obtained from linear regression analysis. Estimated pKB values for propranolol were obtained using the following equation: pKB = –log[antagonist concentration/(CR – 1)], where CR is the concentration ratio of the agonist in the presence and absence of the antagonist.

Functional protocols Concentration–response curves were constructed for two putative β 3-adrenoceptor agonists, CGP12177A and SR 58611A, and for isoproterenol in the absence and presence of various concentrations (0.1 and 1 µmol/L) of the nonselective β1/β 2-antagonist propranolol or of a combination of the selective β1-antagonist CGP20712A and the selective β 2antagonist ICI 118,551. The isoproterenol-induced maximal response was compared with that induced by a maximal dose of papaverine (0.1 mmol/L). When β-antagonists were used, cells were preincubated with the compounds for 5 min, then with isoproterenol for 60 s, then with carbachol for 30 s, and then the reaction was stopped with acrolein.

SMC isolated from the taenia and circular muscle layers of human colon have similar resting cell lengths (85.3 ± 0.9 and 85.9 ± 0.2 µm, respectively) and similar maximal contractions in response to 30 nmol/L carbachol (18.1 ± 0.2 and 16.8 ± 1.0%, respectively). All the β-adrenoceptor agonists tested induced a concentration-dependent relaxation in both types of muscle cells. The effective concentrations, for all agonists, ranged from 10 to 0.1 mmol/L, with supramaximal concentrations inducing a decrease in response.

Detection of mRNA with RT-PCR Total RNA was extracted from both the isolated and dispersed taenia coli, and from circular colonic smooth muscle cell preparations using the RNeasy mini kit (Quiagen, Hilden, Germany). First-strand cDNA for PCR was synthesyzed from 1 µg total RNA, after DNase-I treatment to remove residual genomic DNA, with reverse transcription, using a mixture of random examer and oligo(dT) primers, in accordance with the manufacturer’s standard protocol (Gibco BRL, Gaithersburg, Md.). PCR was carried out in a buffer at pH 8.3, containing, in a 50-µg reaction volume, 2.5 U Taq DNA polymerase and the following (in mmol/L): Tris–HCl, 10; KCl, 50; MgCl2, 1.0; each dNTP, 0.2; each primer, 0.3. Specific human β 3-adrenergic receptor PCR primers, used to amplify a 314-base-pairs PCR product, were 5′-GCTCCGTGGCCTCACGAGAA-3′ and 5′-CCCAACGGCCAGTGGCCAGTCACG-3′. PCR reactions were performed under the following conditions: an initial denaturing step of 10 min at 95 °C; 35 cycles of 50 s denaturation at 95 °C, 50 s annealing at 57 °C, and 60 s extension at 72 °C, with a 5 min final extension at 72 °C. Human brain cDNA was used as a positive control, and negative control reactions were performed without the template in the reaction mixture. Human β-actin primers were used under identical conditions in PCR reactions to verify the integrity of the cDNA template. PCR-product specificity was assessed with Southern blot, using the human-subtype-specific β3-adrenergic receptor primer 5′-CCTAGCCGGGGCCCTGCTGG-3′. Briefly,

Effects of ␤-adrenoceptor antagonists on isoproterenolinduced relaxation The nonselective β-agonist isoproterenol relaxed, in a concentration-dependent manner, both human taenia coli and circular colonic SMC, although intrinsic activity (Table 1) and potency (Table 2) were greater in the taenia coli. Isoproterenol-induced maximal relaxations were similar to those obtained with a maximal dose of papaverine (0.1 mmol/L) (Table1). In the presence of 1 µmol/L of the selective β1-antagonist CGP20712A and 1 µmol/L of the selective β 2-antagonist ICI 118,551, a similar decrease in isoproterenol intrinsic activity was observed in both cell types, with the resulting maximal relaxation of 10 µmol/L (Fig. 1) being approximately 75% of that observed in the absence of β1- and β 2-antagonists (Table 2). In the presence of the two antagonists, the isoproterenol concentration– response curves for taenia coli shifted left. To further analyse the effect of β1- and β 2-blockade on isoproterenolinduced relaxation, isoproterenol effects were studied in taenia coli in the presence of 0.1 and 1 µmol/L of the nonselective β1/β 2 adrenoceptor antagonist propranolol. The isoproterenol concentration–response curve shifted right with propranolol, with mean estimated pKB values of 8.12 ± 0.27 at 0.1 µmol/L and 6.45 ± 0.13 at 1 µmol/L. (Fig. 2). Isoproterenol-induced relaxation was antagonized to a similar extent by the two concentrations of propranolol. Both doses induced a 30%–35% decrease in isoproterenol intrinsic activity, with the maximal relaxation of 41.0 ± 3.7% and

Results

© 2004 NRC Canada

518

Can. J. Physiol. Pharmacol. Vol. 82, 2004

Table 1. Maximal relaxation of smooth muscle cells isolated from taenia coli and circular smooth muscle layers of human sigma. Taenia coli 0.1 mmol/L papaverine 0.1 mmol/L isoproterenol 10 µmol/L isoproterenol + 1 µmol/L [ICI 118551 + CGP20712] 0.1 mmol/L CGP12177A 0.1 mmol/L SR58611A (0.1)

65.4±0.3% 63.8±2.2% 46.7±0.3% 47.9±2.0% 49.7±1.4%

Circular (n (n (n (n (n

= = = = =

3) 4) 5) 3) 3)

54.6±0.3%* (n = 3) 55.2±1.4%* (n = 3) 40.2±0.9%* (n = 3) 44.0±0.3% (n = 3) 45.8±0.3% (n = 3)

Note: Data presented are means ± SEM of n experiments performed in duplicate. *, p < 0.05 taenia coli vs. circular smooth muscle layers.

Table 2. Pharmacological characteristics of β-agonists on smooth muscle cells isolated from taenia coli and circular smooth muscle layers of human sigma. Taenia Coli Isoproterenol Isoproterenol + 1 µmol/L [ICI 118551+CGP20712] (0.1 mmol/L) CGP12177A (0.1 mmol/L) SR58611A

Circular

I.A.

pEC50

I.A.

pEC50

100 77.4±3.0 75.5±3.2 78.4±2.3

6.14±0.04 7.41±0.07 6.22±0.08 6.01±0.05

100 72.9±1.5 79.7±0.5 83.0±0.5

5.93±0.01* 6.32±0.08* 5.90±0.005* 5.95±0.002

Note: Data presented are means ± SEM of at least 3 experiments performed in duplicates. I.A., intrinsic activity relative to isoproterenol (0.1 mmol/L isoproterenol = 100%). pEC50, negative logarithm of concentration producing 50% maximal effect. *, p < 0.05 taenia coli vs. circular smooth muscle layers.

Fig. 1. Dose–response curves for isoproterenol-induced relaxation in the absence (䊉) or presence (䉱) of the selective β 1 -antagonist CGP20712A (1 µmol/L) and the selective β 2 -antagonist ICI 118,551 (1 µmol/L) in taenia coli (left) (n = 5) and circular (right) (n = 3) smooth muscle cells isolated from human sigma. Data are mean ± SEM of n experiments performed in duplicate.

43.4 ± 0.5% in the presence of 0.1 and 1 µmol/L of propranolol, respectively; in the absence propranolol, maximal relaxation was 63.8 ± 2.2%. However, a maximum response to isoproterenol was not achieved in the presence of the antagonist.

Responses to ␤3 -adrenoceptor agonists Isoproterenol-mediated effects were compared with the relaxation induced by the two β3-adrenoceptor agonists. CGP12177A (Fig. 3) and SR 58611A (Fig. 4) dosedependently relaxed both taenia coli and circular colonic © 2004 NRC Canada

Severi et al. Fig. 2. Dose–response curves for isoproterenol-induced relaxation in the absence (䊉) or presence of 0.1 µmol/L (䉱) (n = 4) and 1 µmol/L (䊏) (n = 5) propranolol (PROP) in taenia coli smooth muscle cells isolated from human sigma. Data are mean ± SEM of n experiments performed in duplicate.

SMC, inducing similar maximal responses in both layers. For taenia coli and circular SMC, their mean intrinsic activities were similar to those of isoproterenol in the presence of β1- and β2-antagonists. CGP 12177A was more potent in taenia coli than in circular colonic SMC (Table 2). No differences in potency between the two layers were observed for SR 58611A (Table 2).

519 Fig. 3. Dose–response curves for CGP12177A-induced relaxation in taenia coli (䊉) (n = 3) and circular (䉱) (n = 4) smooth muscle cells isolated from human sigma. Data are mean ± SEM of n experiments performed in duplicate.

Fig. 4. Dose–response curves for SR58611A-induced relaxation in taenia coli (䊉) (n = 3) and circular (䉱) (n = 3) smooth muscle cells isolated from human sigma. Data are mean ± SEM of n experiments performed in duplicate.

Detection of mRNA for ␤3 -adrenoceptors The presence of specific β 3-adrenoceptor mRNAs in taenia coli and circular colonic SMC was determined with RT-PCR. A PCR-product of 314 base pairs was amplified from the cDNA of both taenia coli and circular SMC, with a more marked expression in the former (Fig. 5A). The PCRproduct specificity was confirmed with Southern blot, using a (γ-32P)-labelled β 3-adrenergic receptor internal oligonucleotide (Fig. 5B).

Discussion This study, by evaluating the direct myogenic contribution of β 3-adrenoceptors in mediating human colonic smooth muscle relaxation, supports the functional role of β 3adrenoceptors in both the human taenia coli and circular smooth muscle layers. The presence of β 3-adrenoceptors in human colonic SMC was strongly suggested by the concentration-dependent relaxation exerted by the selective © 2004 NRC Canada

520

Can. J. Physiol. Pharmacol. Vol. 82, 2004

Fig. 5. (A) RT-PCR and (B) Southern blot for β 3-adrenoceptors on taenia coli and circular smooth muscle cells isolated from human sigma. (Circ, circular; T.C., taenia coli; SMC, smooth muscle cells). Positive (Pos) control, human brain cDNA; negative (Neg) control, absence of template in the reaction mixture.

β 3-adrenoceptor agonists CGP12177A and SR58611A and by isoproterenol, even in the presence of selective and nonselective β1- and β 2-adrenoceptors antagonists. The presence β 3-adrenoceptors was also indicated by the molecular detection of β 3-adrenoceptor mRNA both in taenia coli and circular colonic SMC. The functional role β 3-adrenoceptors play in inhibiting human colonic motility has been suggested by the authors of several in vitro functional studies carried out on human muscle-strip preparations from circular and taenia coli colonic SMC (Roberts et al. 1997; De Ponti et al. 1999; Manara et al. 2000). It was assumed that adrenergic agonists inhibited gastrointestinal motility by interacting with a heterogeneous smooth muscle population of β-adrenoceptors, namely β1, β 2, and β 3. The results emerging from our study support this hypothesis. First, by antagonizing β1- and β 2adrenoceptors with the β1-antagonist CGP20712A and the β 2-antagonist ICI 118,551, respectively, isoproterenol intrinsic activity decreasesin colonic SMC isolated from the two layers. Second, the β 3-adrenoceptor agonists CGP12177A and SR58611A have a concentration-dependent relaxing effect on colonic SMC, and their intrinsic activities are comparable to those observed with isoproterenol tested in the presence of CGP20712A and ICI 118,551 antagonists (i.e., when it could not act on both β1- and β2-adrenoceptors). It has been shown that, when tested in animal and human studies in vitro, SR58611A is a selective β3-adrenoceptor agonist (Bianchetti and Manara 1990; Manara et al. 2000), whereas CGP12177A, first described as an antagonist of β1- and β 2adrenoceptors, was demonstrated to display is a highly effective agonist for human β 3-adrenoceptors (Wilson et al. 1996). More evidence of the presence of a heterogeneous receptor population of β-adrenoceptors subtypes has been suggested, in studies on taenia coli, by the propranolol antagonism of isoproterenol-induced relaxation. First, the estimated affinity of propranolol was somewhat lower than what would be expected for β1- and β 2-adrenoceptors (Bianchetti

and Manara 1990); this is consistent with the involvement of β 3 in isoproterenol-induced relaxation. Furthermore, on isoproterenol concentration curves, the fact that the shift lessens with progressively higher concentrations of propranolol suggests the presence of a heterogeneous β-adrenoceptor population (Roberts et al. 1999), as does the noncompetitive nature of isoproterenol antagonism with either propranolol or CGP20712A and ICI 118551 (Kelly et al. 1998, Manara et al. 2000). Propranolol, besides interacting with β1- and β 2adrenoceptors, also appears to reduce β3 activity; this reduction is suggested by the fact that, in the presence of propranolol, the decrease in isoproterenol intrinsic activity is greater than that observed with ICI 118551 and CGP20712A. The effect of propranolol on β 3-adrenoceptors was observed in the rat distal colon (McKean and MacDonald 1995). The experimental conditions (e.g., time of antagonist preincubation) probably did not influence the effects of the antagonists, because SMC preparations eliminate diffusion barriers, allowing cells to rapidly reach an equilibrium with membrane receptors, eliminating the need for the longer preincubations required by muscle strips (Severi et al. 1988, 1989). Both isoproterenol and CGP12177A were found to be more potent in human taenia coli than in circular colonic SMC. The greater responsiveness of taenia coli to β 3selective and nonselective agonists may be related to the different expression of adrenoceptors in the two layers. This study showed that expression of β 3-adrenoceptor mRNA in taenia coli SMC was more marked; circular colonic SMC had low, but detectable, β 3-mRNA. Previous studies have only detected β 3-adrenoceptor mRNA in combined longitudinal/circular smooth muscle preparations, but have failed to establish the real contribution of the circular smooth muscle layer to the expression observed (Roberts et al. 1997). In addition to the different expression of β 3adrenoceptors in the two muscle layers, other mechanisms should be considered. No differences in potency were ob© 2004 NRC Canada

Severi et al.

served with SR58611A, the other β 3-adrenoceptor agonist tested, and the antagonism of β1-/β 2-receptor effects appears to increase β 3-mediated effects, as suggested by the leftward shift of isoproterenol concentration–response curves for taenia coli SMC in the presence of CGP20712A and ICI 118,551. Recent studies have shown that the pharmacological and functional properties of β-adrenoceptors are different when the coexpression of multiple receptors subtypes is considered, than whenthe expression of a single subtype is looked at (Lavoie and Hebert 2003). In addition to the common coupling of β-adrenoceptor to Gs, the G-protein that stimulates adenylate cyclase, a preferential interaction of a given β-adrenoceptor subtype with one (or several) additional G-protein(s) has been reported (Emorine et al. 1991; Robillard et al. 2000). These specific properties of individual β-adrenoceptor subtypes may allow each subtype to preferentially regulate individual enzymatic pathways, even when other β-adrenoceptor subtypes are expressed by the same cells, and blocking one or more of these subtypes may enhance a specific transduction pathway. Such a mechanism could explain the increase in the potency of isoproterenol in the presence of CGP20712A and ICI 118,551 that was observed in taenia coli SMC, and the higher potency, in taenia coli, of CGP12177A, a β 3-agonist with β1-/β 2-antagonist activities. Whereas the role of β 3-adrenoceptors in mediating relaxation of human taenia coli has been well established (Roberts et al. 1997; Kelly et al. 1998; Manara et al. 2000), studies on circular muscle strips, even if supporting the functional importance of β3-adrenoceptors, have shown conflicting results concerning agonist activity on this muscle layer (De Ponti et al. 1999; Manara et al. 2000). On the other hand, our study supports the role of β 3-adrenoceptors in mediating circular colonic SMC relaxation. Isoproterenol activity was only partly inhibited by β1- and β 2-antagonists, the two β 3-agonists CGP12177A and SR58611A relaxed circular colonic SMC in a dose-dependent manner, and β 3adrenoceptor mRNA was amplified from the cDNA. The discrepancies observed in the relaxation induced by β 3adrenoceptors agonists in circular smooth muscle preparations (De Ponti et al. 1999; Manara et al. 2000) might be due to tissue differences and (or) the heterogeneity of the patients (i.e., different sympathetic activity). The coexistence, in each preparation, of a different relative abundance and functional importance of the β-adrenoceptor subtypes (β1, β 2, and β 3) could affect agonist responses. Another factor that might explain the differences in agonist activity in gastrointestinal smooth muscle is receptor desensitization (Severi et al. 1999). Muscle cells, in intact strips, might be exposed to a background of neurohumoral agents, which could interact with membrane receptors, causing their desensitization. This phenomenon could be of particular functional importance in a tissue in which the presence of β 3-adrenoceptors is relatively low. Immunohistochemical studies have shown more β 3-adrenoceptors in the longitudinal than in the circular colonic muscle layer (Anthony et al. 1998), and our findings on mRNA expression confirm a more marked expression in taenia coli than in circular colonic SMC. The process of cell dispersion, such as that used in our study, not only eliminates the neurohumoral environment, possibly allowing β 3adrenoceptors to reappear, but also offers an advantage to

521

experiments performed with morphologic homogeneous cell types in which environmental influences, such as diffusion barriers, are not present (Makhlouf 1987). In conclusion, all three currently recognized βadrenoceptors subtypes variably account for the relaxation of human taenia coli and circular colonic SMC. There is strong functional evidence of β 3-adrenoceptor involvement in human colon motility and, hence, of a potential target for the pharmacological control of gut motility.

Acknowledgements This work was supported by an interuniversity cofinanced program (No. 9906218982) from Italian Ministry for University and Technological Research (MURST) (1999). The authors are grateful to Mrs. Marian Shields for her help with the language.

References Anthony, A., Schepelmann, S., Guillaume, J-L., Strosberg, A.D., Dhillon, A.P., Pounder, R.E., and Wakefield, A.J. 1998. Localization of the β (beta)3-adrenoceptor in the human gastrointestinal tract: an immunochemical study. Aliment. Pharmacol. Ther. 12: 519- 522. Bardou, M., Dousset, B., Deneux–Tharaux, C., Smadja, C., Naline, E., Chaput, J-C., Naveau. S., Manara, L., Croci, T., and Advenier, C. 1998. In vitro inhibition of human colonic motility with SR59119A and SR59104A: evidence of β 3-adrenoceptormediated effect. Eur. J. Pharmacol. 353: 281–287. Berkowitz, D.E., Nardone, N.A., Smiley, R.M., Price, D.T., Kreutter, D.K., Fremeau, R.T., and Schwinn, D.A. 1995. Distribution of β 3-adrenoceptor mRNA tissues. Eur. J. Pharmacol. 289: 223–228. Bianchetti, A., and Manara, L. 1990. In vitro inhibition of intestinal motility by phenylethanolaminotetralines: evidence of atypical β-adrenoceptors in rat colon. Br. J. Pharmacol. 100: 831– 839. Bitar, K.N., and Makhlouf, G.M. 1982a. Receptors on smooth muscle cells: characterization by contraction and specific agonists. Am. J. Physiol. 242: G400–G407. Bitar, K.N., and Makhlouf, G.M. 1982b. Relaxation of isolated gastric smooth muscle cells by vasoactive intestinal peptide. Science, (Wash., D.C.) 261: 531–533. Burnstock, G., and Wong, H. 1981. Systemic pharmacology of adrenergic agonists and antagonists: effects on the digestive system. In Handbook of experimental pharmacology. Edited by L. Szekeres. Springer-Verlag, New York. pp. 129–159. De Ponti, F., Gibelli, G., Croci, T., Arcidiaco, M., Crema, F., and Manara, L. 1996. Functional evidence of atypical β 3adrenoceptor in the human colon using the β 3-selective adrenoceptor antagonist SR59230A. Br. J. Pharmacol. 117: 1374–1376. De Ponti, F., Mondini, C., Gibelli., G., Crema., F., and Frigo., G. 1999. Atypical β-adrenoceptors mediating relaxation in the human colon: functional evidence for β 3 -rather than β 4 adrenoceptors. Pharmacol. Res. 39: 345–348. Ek, B.A., Bjellin, L.A.C., and Lundgren, B.T. 1986. Betaadrenergic control of motility in the rat colon. I. Evidence for functional separation of the beta1- and beta2-adrenoceptor mediated inhibition of colon activity. Gastroenterology, 90: 400–407. Emorine, L.J., Feve, B., Pairault, J., Briend-Sutren, M.-M., Marullo, S., Delavier-Klutchko, C., and Strosberg, D.A. 1991. © 2004 NRC Canada

522 Structural basis for functional diversity of β 1 -, β 2 - and β 3 adrenergic receptors. Biochem. Pharmacol. 41: 853–859. Furness, J.B., and Costa, M. 1974. The adrenergic innervation of the gastrointestinal tract. Ergeb. Physiol. Biol. Chem. Exp. Pharmacol. 69: 1–15. Hallerback, B., Carlsen, E., Carlsson, K., Enkvist, C., Glise, H., Haffner, J., Innes, R., and Kirno, K. 1987. Beta-adrenoceptor blockade in the treatment of postoperative adynamic ileus. Scand. J. Gastroenterol. 22: 149–155. Kelly, J., Sennit, M.V., Stock, M.J., and Arch, R.S. 1998. Functional evidence for atypical β-adrenoceptors in human isolated taenia coli. Pharmacol. Rev. Commun. 10: 143–152. Koike, K., Takayanagi, I., Ichino, T., Koshikawa, H., and Nagatomo, T. 1997. β 3 -Adrenoceptor mediated relaxation of guinea pig taenia caecum by BRL 37344A and BRL 35135A. Eur. J. Pharmacol. 334: 217–221. Krief, S., Lönnqvisit, F., Raibault, S., Baude, B., Van Spronser, A., Arner, P., Strosberg, D., Riquier, D., and Emorine, L.J. 1993. Tissue distribution of β 3 -adrenergic receptor mRNA in man. J. Clin. Invest. 91: 344–349. Landi, M., Croci, T., and Manara, L. 1993. Similar atypical βadrenergic receptors mediate in vitro rat adipocyte lipolysis and colonic motility inhibition. Life Sci. 53: 297–302. Lavoie, C., and Hebert, T.E. 2003. Pharmacological characterization of putative β 1 -β 2 -adrenergic receptor heterodimers. Can. J. Physiol. Pharmacol. 81: 186–195. Lezama, E.J., Konkar, A.A., Salazar-Bookaman, M.M., Miller, D.D., and Feller, D.R. 1996. Pharmacological study of atypical β-adrenoceptors in rat esophageal smooth muscle. Eur. J. Pharmacol. 308: 69–80. Lindgren, S., Lilja, B., Rosen, I., and Sundkvist, G. 1991. Disturbed autonomic nerve function in patients with Crohn’s disease. Scand. J. Gastroenterol. 26: 361–366. Luckensmeyer, G.B., and Keast J.R. 1998. Activation of α- and βadrenoceptors by symphathetic nerve stimulation in the large intestine of the rat. J. Physiol. (Lond.), 510(2): 549–561. Lyrenas, E. 1985. Beta adrenergic influence on esophageal and colonic motility in man. Scand. J. Gastroenterol. 116: 1–48. Makhlouf, G.M. 1987. Isolated smooth muscle cells of the gut. In Physiology of the gastrointestinal tract. Edited by L.R. Johnson. Raven Press, New York. pp. 555–587. Manara, L., Croci, T., and Landi, M. 1995. β 3 adrenoceptors and intestinal motility. Fundam. Clin. Pharmacol. 9: 332–342. Manara, L., Croci, T., Aureggi, G., Guagnini, F., Maffrand, J.-P., Le Fur, G., Mekenge, S., and Ferla, G. 2000. Functional assessment of β-adrenoceptor subtypes in human colonic circular and longitudinal (taenia coli) smooth muscle. Gut, 47: 337–342.

Can. J. Physiol. Pharmacol. Vol. 82, 2004 McKean, J., and MacDonald, A. 1995. Contribution of βadrenoceptor subtypes to responses to isoprenaline in rat isolated distal colon. J. Pharm. Pharmacol. 47: 388–391. Norberg, K.-A. 1964. Adrenergic innervation of the intestinal wall studied by fluorescence microscopy. Internat. J. Neuropharmacol. 3: 379–382. Oostendorp, J., Preitner, F., Moffatt, J., Jimenez, M., Giacobino, J.P., Molenaar, P., and Kaumann, A.J. 2000. Contribution of βadrenoceptor subtypes to relaxation of colon and oesophagus and pacemaker activity of ureter in wild-type and β 3 adrenoceptor knockout mice. Br. J. Pharmacol. 130: 747–758. Roberts, S.J., Papaioannou, M., Evans, B.A., and Summer, R.J. 1997. Functional and molecular evidence for β 1 -, β 2 - and β 3 adrenoceptors in human colon. Br. J. Pharmacol. 102: 1527– 1535. Roberts, S.J., Papaioannou, M., Evans, B.A., and Summer, R.J. 1999. Characterization of β-adrenoceptor-mediated smooth muscle relaxation and the detection of mRNA for β 1 -, β 2 -, β 3 adrenoceptors in rat ileum. Br. J. Pharmacol. 127: 949–961. Robillard, L., Ethier, N., Lachance, M., and Hebert, T.E. 2000. Gbetagamma subunit combinations differentially modulate receptor and effector coupling in vivo. Cell. Signal. 12: 673–682. Severi, C., Grider, J.R., and Makhlouf, G.M. 1988. Characterization of opioid receptors on isolated canine gallbladder smooth muscle cells. Life Sci. 42: 2373–2380. Severi, C., Coy, D.H., Jensen, R.T., Boschero, L., Anania, M.C., and Delle Fave, G. 1989. Pharmacological characterization of [Leu13-ϕCH2NH-Leu14]-bombesin as a specific bombesin receptor antagonist on isolated smooth muscle cells. J. Pharmacol. Exp.Ther. 251: 713–717. Severi, C., Carnicelli., V., Di Giulio, A., Romano, G., Bozzi, A., Oratore, A., Strom, R., and Delle Fave, G. 1999. Progression from homologous to heterologous desensitization of contraction in gastric smooth muscle cells. J. Pharmacol. Exp. Ther. 288: 389–398. Summers, R.J. 1999. β 3 -Adrenoceptors: their role and regulation in the gastrointestinal tract. Proc. West. Pharmacol. Soc. 42: 115– 117. van der Vliet, A., Rademarker, B., and Bast, A. 1990. A beta adrenoceptor with atypical characteristics is involved in the relaxation of the rat small intestine. J. Pharmacol. Exp. Ther. 253: 218–226. Wilson, S., Chambers, J.K., Park, J.E., Ladurner, A., Cronk, D.W., Chapman, C.G., Kallender, H., Browne, M.J., Murphy, G.J., and Young, P.W. 1996. Agonist potency at the cloned β 3 adrenoceptor depends on receptor expression level and nature of the assay. J. Pharmacol. Exp. Ther. 279: 214–221.

© 2004 NRC Canada