Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

It is well known that postjunctional -adrenoceptors (-ARs) mediate the inhibitory effects of sympathetic nerve stimulation in different smooth muscles including those of the gastrointestinal tract (Manara et al., 1995b; Gauthier et al., 2000). The intestinal -AR was originally described as a ₁- and ₂-AR (Lands et al., 1967). Further studies with gastrointestinal preparations from several species established the relaxant effect of classical -AR (₁ and ₂) agonists (Bennett, 1965; Hedges and Turner, 1969; De Ponti et al., 1996a). Subsequently, studies investigating -ARs in gastrointestinal smooth muscle from several species demonstrated relaxation responses that were resistant to propranolol and displayed lower affinity to other conventional -AR antagonists (Arch and Kaumann, 1993; Goldberg and Frishman, 1995; Strosberg, 1997; Manara et al., 2000). This finding, along with the emergence of a new class of -AR agonists described first in adipocytes (Feve et al., 1991), suggested the presence of an "atypical" class of -ARs. In 1989, Emorine et al. (1989) cloned and sequenced ₃-AR and found that it shared the pharmacological characteristics of the atypical -AR.

The ₃-AR has been found in a variety of mammalian tissues (Berkowitz et al., 1995) including white and brown adipocytes (Muzzin et al., 1991), trachea (Webber and Stock, 1992), heart (Kaumann and Molenaar, 1996; Gauthier et al., 2000), gastrointestinal tract (De Ponti et al., 1995; Bardou et al., 1998), and urinary tract (Tomiyama et al., 1998). In the GI tract, recent studies have focused on the ability of ₃-AR specific agonists to cause relaxation in a number of different smooth muscle tissues including rat ileum, jejunum, colon, guinea pig ileum, and duodenum (Manara et al., 1995b). One of the problems in delineating the pharmacology of -ARs in the gastrointestinal tract has been the lack of subtype-selective agonists and antagonists, especially those for ₃-AR. Recent in vivo studies have demonstrated the selective, potent, and prolonged relaxant effect of CL 316,243, a selective ₃-AR agonist, on the sphincteric smooth muscles of the opossum lower esophageal sphincter (DiMarino et al., 2002), without the significant systemic cardiovascular side effects that are associated with ₁- and ₂-AR agonists.

In the past few years, ₃-agonists have emerged as potential therapeutic agents for several gastrointestinal motility disorders including irritable bowel syndrome (Scarpignato and Pelosini, 1999). Anorectal dysfunctions such as Hirschsprung's disease, constipation, anal fissures, and hemorrhoids may also be associated with either hypertensive IAS or failure of sphincteric relaxation in response to the rectoanal inhibitory reflex (Azpiroz and Whitehead, 2002). Characterization of neurohumoral receptors that mediate selective, potent, and prolonged relaxation of IAS and other GI smooth muscles without untoward systemic effects will be of considerable interest in the treatment of anorectal and other GI motility disorders.

The present investigation was carried out to characterize -AR in the gastrointestinal tonic smooth muscle of the IAS by comprehensive studies using a combination of classical pharmacology, receptor binding, and molecular biology approaches.

The aim of the present study is to determine the presence of and characterize the -AR subtypes involved in mediating relaxation of the IAS smooth muscle. We used selective agonists and antagonists to determine the receptor binding profiles of each -AR subtype. The presence of membrane bound -AR and mRNA encoding for the three -AR subtypes was determined through Western blot studies and reverse transcription-polymerase chain reaction (RT-PCR) analysis, respectively.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Preparation of Smooth Muscle Strips. Adult male opossums (Didelphis virginiana) weighing 2.5 to 3.5 kg were anesthetized with sodium pentobarbital (50 mg/kg i.p.). Laparotomy was performed, and a part of the rectum along with the anal canal was removed using sharp dissection. The IAS was identified by manometry as high-pressure zone and marked by means of sutures in situ. The animals were sacrificed by exsanguinations; the anorectal region was then dissected out and transferred immediately to oxygenated (95% O₂ + 5% CO₂) Krebs' physiological solution of the following composition: 118.07 mM NaCl, 4.69 mM KCl, 2.52 mM CaCl₂, 1.16 mM MgSO₄, 1.01 mM NaH₂PO₄, 25 mM NaHCO₃, and 11.10 mM glucose. A longitudinal incision along the length of isolated anorectal region was made, and the tissue was pinned flat in a Sylgard (Dow Corning Corp., Midland, MI)-coated Petri dish. Once the lumen was fully exposed, the mucosa and submucosa were removed carefully by sharp dissection. The tissue was then turned on the serosal side, and all extraneous tissue including the outer longitudinal muscle was removed. Circular smooth muscle strips of the IAS (approximately 1 × 10 mm) were prepared and tied on either end using 3-0 silk suture in preparation for measurement of isometric tension.

Measurement of Isometric Tension. The smooth muscle strips were transferred to 2-ml muscle baths (Radnoti Glass Technology, Inc., Monrovia, CA) containing oxygenated Krebs' solution at 35°C. One end of the muscle strip was anchored at the bottom of the muscle bath while the other end was connected to a force transducer (model FT03; Grass Instruments, Quincy, MA). Isometric tension was recorded by the PowerLab/8SP data acquisition system using Chart 4.1.2 (ADInstruments, Grand Junction, CO). Each smooth muscle strip was initially stretched to a tension of 0.7 g. The muscle strips were then given at least an hour to equilibrate, during which time they were washed with Krebs' solution every 15 min. Only smooth muscle strips that developed spontaneous tone and responded to electrical field stimulation were used in this study. The changes in tension from various drugs were expressed as the percent maximal relaxation achieved by 50 mM EGTA, at the end of each experiment. Each smooth muscle served as its own control.

Drug Responses. To determine the concentration-response curves (CRCs) with ₁-, ₂-, and ₃-AR agonists on the basal tone of the IAS smooth muscles, xamoterol, procaterol, and ZD 7114 [(S)-4-[2-hydroxy-3-phenoxypropylaminoethoxy]-N-(2-methoxyethyl)phenoxyacetamide], respectively, were added to the muscle bath in cumulative concentrations (Rattan and Moummi, 1989). Successive concentrations of the agonists were not added until the response of the previous concentration stabilized. When no effect was observed, 10 minutes were allowed between additions of different concentrations. In preliminary studies, when a single concentration was used, we noted that this was an appropriate time to gauge the maximal effect of a given concentration of the agonist. No difference in the results occurred with longer exposures. To determine the effects of ₁-, ₂-, and ₃-AR antagonists, CGP 20712A [(±)-2-hydroxy-5-[2-[[2-hydroxy-3-[4-[1-methyl-4-(trifluoromethyl)-1H-imidazol-2-yl]phenoxy]propyl]amino]ethoxy]-benzamide methanesulfonate salt], ICI 118551, and SR 59230A [1-(2-ethylphenoxy)-3-[[(1S)-1,2,3,4-tetrahydro-1-naphthalenyl]amino]-(2S)-2-propanol hydrochloride], respectively (in concentrations ranging from 1 × 10⁸ to 1 × 10⁶ M), were added 30 min before obtaining the CRC of the test agonist.

-Adrenoceptor (-AR) Analysis by Western Blot. Western blot analysis of ₁-, ₂-, and ₃-AR in the IAS and rectum of the opossum was performed according to the protocol of Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Circular smooth muscles tissues of the IAS and rectum were cut into small pieces (2 × 2 mm cubes) and rapidly homogenized in 3 ml of boiling lysis buffer (1% SDS, 1.0 mM sodium orthovanadate, 10 mM Tris, pH 7.4) and then put into the microwave for 10 s. The homogenates were centrifuged (16,000g, 4°C) for 15 min. The pellet obtained was dissolved in Krebs' buffer (composition already described) containing 1 mM EDTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride (combined pH of 7.6). The protein contents were determined by the method described by Lowry et al. (1951) using bovine serum albumin as the standard.

Membrane Preparation for Receptor Binding Studies. The circular smooth muscle of the IAS was dissected free by the aforementioned procedure and placed immediately in ice-cold Krebs' buffer (composition already described above) containing 1 mM EDTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride (combined pH of 7.6). The IAS was minced with scissors and homogenized in 5 volumes of ice-cold TED buffer (20 mM Tris Cl, 1 mM EDTA, 1 mM dithioreitol, pH 8) by the use of a Tekmar Tissuemizer (Tekmar-Dohrmann, Mason, OH) for 15 s. The homogenates were centrifuged at 100,000g for 1 h at 4°C. The supernatant was filtered through a 500-µm Nitex mesh. The pellets were resuspended in cold Krebs' buffer (pH 7.6) and stored at 80°C until used. Protein content was determined by the method of Lowry et al. (1951).

Radioligand Binding Studies. The radioligand ()-3-[¹²⁵I]iodocyanopindolol ([¹²⁵I]CYP; Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK) was used for identifying -AR. For equilibrium determination, membranes at a protein concentration of 40 µg per tube were incubated with [¹²⁵I]CYP (specific activity 2000 Ci/mmol) for 0, 15, 30, 45, 60, 90, 120, 150, and 180 min. The experiments were carried out in the presence or absence of 100 µM propranolol (a nonselective -AR antagonist). The incubation mixture was composed of 50 mM Tris HCl buffer, pH 7.4, containing 10 mM MgCl₂ and 1 mM EDTA in a final volume of 250 µl. A time course (using the above-mentioned time points) was carried out in duplicate at 35°C to determine the optimal time needed for equilibrium. The incubation was terminated by rapid filtration through Whatman GF/C glass-fiber filters (24-mm circles) (Whatman, Clifton, NJ) using a 1225 sampling manifold (Millipore Corp., Bedford, MA), followed by washing three times with 5 ml of ice-cold 25 mM Tris HCl buffer, pH 7.4. The filters were counted in the Auto-Gamma Counting System (model 5550; PerkinElmer Life Sciences, Boston, MA) at an efficiency of 80%. Specific binding was calculated by subtracting nonspecific binding from total binding.

Isolation and Quantification of Total RNA. Tissue specimens from the circular smooth muscle of the IAS were carefully dissected and homogenized as described under Membrane Preparation for Receptor Binding Studies. Total RNA was extracted from the tissue homogenate using the TRI reagent (Molecular Research Center, Cincinnati, OH) protocol based on the method of Chomczynski and Sacchi (1987). RNA samples were then dissolved in diethylpyrocarbonate (DEPC)-treated water (pH 7.5). The optical density (OD) of each sample was determined by a UV-visible spectrophotometer (Amersham Biosciences) at a wavelength of 260 nm (₂₆₀). The yield and quality of the RNA were assessed by measuring the OD ₂₆₀/OD ₂₈₀ ratio.

Preparation and Amplification of cDNA Encoding ₁-, ₂-, and ₃-ARs (RT-PCR Analysis). RNA samples of 2 µl (1 µg) that were of acceptable quality were used as templates for the synthesis of cDNA. Primers for ₁-, ₂-, and ₃-AR, and -actin (internal standard), based on the previous report (Dincer, 2002), were synthesized by Thomas Jefferson University facilities (Kimmel Cancer Institute, Nucleic Acid Facility). The sequence and accession numbers listed in Table 1 are based on published sequences in the National Center for Biotechnology Information GenBank database (http://www3.ncbi.nlm.nih.gov/entrez). cDNAs were synthesized by reverse transcription of 1.0 µg of each total RNA. The reaction mixture consisted of 10× reverse transcription buffer, deoxynucleoside-5'-triphosphates (20 mM), MgCl₂ (25 mM), 18 U of RNasin ribonuclease inhibitor, and 20 U of AMV reverse transcriptase in a total volume of 20 µl. The contents of reaction mixture were purchased from Promega (Madison, WI). After a brief centrifugation, the reaction mixtures were incubated at 42°C for 45 min and then at 95°C for 5 min.

View this table: [in this window] [in a new window]   TABLE 1 Primers used in RT-PCRs for amplification of mRNA encoding -AR and -actin in IAS smooth muscle

Drugs and Chemicals. SR 59230A hydrochloride, propranolol hydrochloride [(±)-1-isopropylamino-3-(1-naphthyloxy)-2-propanol hydrochloride], CGP 20712A (methanesulfonate salt), dimethyl sulfoxide (DMSO), and EGTA (ethylene-bis(oxyethylenenitrilo)tetraacetic acid) were purchased from Sigma-Aldrich (St. Louis, MO). Xamoterol hemifumarate [1-(4-hydroxyphenoxy)-3-[2-(4-morpholinocarboxamido)ethylamino]-2-propanol], ICI 118,551 hydrochloride [(±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol], procaterol hydrochloride [(±)-erythro-8-hydroxy-5-[1-hydroxy-2-(isopropylamino)butyl]carbostyril], and ZD 7114 hydrochloride were purchased from Tocris Cookson (Ballwin, MO). [¹²⁵I]CYP was purchased from Amersham Biosciences UK Ltd.

2

Data Analysis. The fall in basal tension of the IAS smooth muscle following administration of agonists was expressed as the percentage of maximal relaxation as explained above. The results were expressed as means ± S.E. of different experiments. The statistical significance between different groups was determined by analysis of variance and by paired or unpaired t test. A p value smaller than 0.05 was considered significant. Agonist potencies, pA₂ of antagonists, and receptor binding data (B_max, K_d, and K_i) were calculated using GraphPad Prism software. pA₂ values were calculated based on the earlier method (Arunlakshana and Schild, 1959).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of ZD 7114 on the Basal Tone of IAS Smooth Muscle. The ₃-AR agonist ZD 7114 (formerly ICI D7114) (Growcott et al., 1993b) produced a concentration-dependent fall in the basal tension of the IAS smooth muscle (Fig. 1A) with an EC₅₀ value of 5.30 × 10⁸ M (n = 8-10). The concentration causing maximal relaxation (EC_max) was 1 × 10⁶ M. The maximal relaxation in different experiments ranged from 80.7 to 88.5%. The selective ₃-AR antagonist SR 59230A (De Ponti et al., 1996b) significantly attenuated the relaxant response to ZD 7114 in a concentration-dependent manner (p < 0.05; n = 5-8; Fig. 1A). A Schild plot produced a line with a slope of 0.90 ± 0.15 (Fig. 1B) and a corresponding pA₂ value of 7.8 ± 0.24. 

View larger version (14K): [in this window] [in a new window]   Fig. 1.   A, effect of 3-agonist ZD 7114 on the basal tone of the IAS (shown as percent maximal fall in basal IAS tone) in the absence (control) and presence of the 3-selective antagonist SR 59230A. SR 592330A causes significant and concentration-dependent rightward shift in the control CRC of ZD 7114 (, p < 0.05; n = 5-8). The values represent mean ± S.E. B, Schild plot of different concentrations of SR 59230A versus log(r  1) of ZD 7114. The pA2 value of SR 59230A in antagonizing ZD 7114-induced relaxation of the IAS smooth muscle is 7.8. C, influence of 2- and 1-AR antagonists ICI 118,551 and CGP 20712A, respectively, on percent maximal fall in basal IAS tone. Data show that the 3-AR agonist-mediated fall in the basal tone of the IAS smooth muscle was not significantly modified by 2- and 1-AR antagonists (p > 0.05; n = 5-8).

7

7

View this table: [in this window] [in a new window]   TABLE 2 Comparison of potencies of 3-, 1-, and 1-AR agonists and their respective antagonists SR 59230A, ICI 118551, and CGP 201712 in the IAS smooth muscle

Effect of Procaterol on the Basal Tone of IAS Smooth Muscle. Procaterol, a ₂-AR selective agonist (Kotsonis and Majewski, 1994) produced a concentration-dependent fall in basal tension of the IAS smooth muscle with an EC₅₀ value of 2.51 × 10⁸ M (n = 5-8) (Fig. 2A). The concentration causing maximal relaxation (EC_max) was 3 × 10⁶ M. The maximal relaxation in different experiments ranged from 79.1 to 83.7%. The selective ₂-AR antagonist ICI 118551 (Bilski et al., 1983) significantly attenuated the relaxant response to ZD 7114 in a concentration-dependent manner (p < 0.05; n = 5-8; Fig. 2A). A Schild plot produced a line with a slope of 0.88 ± 0.07 (Fig. 2B) and a corresponding pA₂ value of 7.70 ± 0.31. 

View larger version (14K): [in this window] [in a new window]   Fig. 2.   A, CRC showing IAS smooth muscle relaxation by procaterol (2-agonist) before and after a selective 2-antagonist, ICI 118551. As shown, ICI 118551 causes a significant and concentration-dependent attenuation in the CRC of procaterol (, p < 0.05; n = 5-8). B, Schild plot of different concentrations of ICI 118551 versus log(r  1) of procaterol. The pA2 value for ICI 118551 in antagonizing procaterol-induced relaxation of the IAS smooth muscle is 7.7. C, 1-AR (CGP 20712A) and 3-AR (SR 59230A) antagonists, in contrast, have no significant effect on the IAS smooth muscle relaxation caused by procaterol (p > 0.05; n = 5-8).

7

7

Effect of Xamoterol on the Basal Tone of IAS Smooth Muscle. The ₁-AR agonist xamoterol (Malta et al., 1985) produced a concentration-dependent fall in the basal tension of the IAS smooth muscle (Fig. 3A) with an EC₅₀ value of 1.02 × 10⁷ M (n = 5-8). The concentration causing maximal relaxation (EC_max) was 3 × 10⁶ M. The maximal relaxation in different experiments ranged from 71.5 to 78.7%. The selective ₁-AR antagonist CGP 20712A (Dooley et al., 1986) caused a significant shift in the CRC of xamoterol in a concentration-dependent manner (p < 0.05; n = 5-8; Fig. 3A). A Schild plot produced a line with a slope of 0.82 ± 0.08 (Fig. 3B) and a corresponding pA₂ value of 7.12 ± 0.18. 

View larger version (13K): [in this window] [in a new window]   Fig. 3.   CRC showing percent maximal fall in basal IAS tone with xamoterol (1-agonist) before and after different concentrations of CGP 20712A (1-antagonist). Data show that CGP 20712A (with the exception of 1 × 108 M) causes a significant and concentration-dependent inhibition of the IAS smooth muscle relaxation by xamoterol (, p < 0.05; n = 5-8). B, Schild plot of different concentrations of CGP 20712A versus log(r  1) of xamoterol. The pA2 value for SR 59230A in antagonizing xamoterol-induced relaxation of the IAS smooth muscle is 7.12. C, influence of 2- and 3-AR antagonists ICI 118,551 and SR 59230A, respectively, on percent maximal fall in basal IAS tone by xamoterol. Data show that the 1-AR agonist-mediated fall in the basal tone of the IAS smooth muscle was not significantly modified by SR 59230A (p > 0.05; n = 5-8) but was modestly antagonized by ICI 118551 (, p < 0.05; n = 5-8).

7

7

7

Receptor Binding Studies on -ARs in IAS Smooth Muscle. To characterize and determine the levels of -ARs in the IAS, we conducted radioligand binding studies with [¹²⁵I]CYP. Based on reports that [¹²⁵I]CYP has a significantly lower affinity for ₁/₂-AR than for ₃-AR (Dunigan et al., 2000; Kohout et al., 2001), we investigated the binding profiles of the three -AR subtypes in the IAS. Initially, to determine the appropriate time need for the equilibrium, a time course was plotted. [¹²⁵I]CYP specifically bound to membrane preparations of the IAS in a time-dependent fashion, with equilibrium achieved at 90 min (35°C), and remained constant for 180 min (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 4.   A, binding of [125I]iodocyanopindolol to -AR in IAS membrane preparations. Representative equilibrium binding curves show saturable and specific binding to -AR. Membrane preparations of opossum IAS (40 µg/tube) were incubated with increasing concentrations of radioligand. Specific binding was quantified as described under Materials and Methods. B, sigmoid representation of the data illustrates the binding of [125I]CYP over large concentration ranges from the high-affinity site (picomolar) to the low affinity site (nanomolar). C, Scatchard transformation of the data reveals a curvilinear plot demonstrating the presence of two binding sites. Linear regression analysis resulted in a two-site binding model characterized by a high- (Kd = 96 pM [125I]CYP; Bmax = 12.5 fmol/mg protein) and low- (Kd = 1.96 nM [125I]CYP; Bmax = 58.7 fmol/mg protein) affinity site.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Competition curves of -AR-selective antagonists for [125I]CYP binding to opossum IAS membrane preparations at a high-affinity site. Membrane preparations of opossum IAS (40 µg/tube) were incubated with 66 pM [125I]CYP and increasing concentrations of subtype-selective antagonists. The figure represents means ± S.E. from four displacement experiments. The corresponding inhibition constants were obtained by the Cheng-Prusoff equation and are listed in Table 3.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Competition curves of selective -AR antagonists for [125I]CYP binding to opossum IAS membrane preparations at a low-affinity site. Membrane preparations of opossum IAS (40 µg/tube) were incubated with 1.61 nM [125I]CYP and increasing concentrations of subtype-selective antagonists. The figure represents means ± S.E. from four displacement experiments.

View this table: [in this window] [in a new window]   TABLE 3 Ki values of -subtype-selective antagonists from competition binding experiments with [125I]CYP

Determination of ₁-, ₂-, and ₃-AR Membrane Protein in the IAS and Rectum. To identify and quantify -AR protein expression in the rectum and IAS, the membrane preparations were fractionated by SDS-polyacrylamide gel electrophoresis and subjected to Western blotting by primary antibodies specific to each -AR subtype (see Materials and Methods). All three subtypes of -AR were found to be present in the rectum and IAS membranes as shown by the representative blots in Fig. 7. The blots demonstrate the relative distribution of membrane receptor proteins for ₁-AR, (63 kDa), ₂-AR (68 kDa), and ₃-AR (65 kDa) in these tissues. Data suggest that the distribution of the three subtypes of membrane -AR in these tissues was similar (p > 0.05; Fig. 7).

View larger version (22K): [in this window] [in a new window]   Fig. 7.   A, Western blot analysis of 1-AR, 2-AR, and 3-AR expression in the plasma membrane of rectum and IAS demonstrating relative distribution of expected size protein for 1-AR (63 kDa), 2-AR (68 kDa), and 3-AR (65 kDa). The membrane protein (40 µg/well) was run on a 7.5% SDS-polyacrylamide gel, electrophoresed for 60 min, transferred to nitrocellulose membranes, and probed by isoform-specific antibodies for each -AR protein. B, densitometric analysis shows equal distribution of different -ARs in the IAS and rectum. Data are means ± S.E. of four experiments.

Detection of -AR mRNA in the IAS Using RT-PCR. RT-PCR amplification was used to detect ₁-, ₂-, and ₃-AR, and -actin mRNA in the circular smooth muscle layer of the IAS. To ensure that the PCR products were exclusively derived from mRNA, total RNA samples were treated with DNase to eliminate genomic DNA. As shown in Fig. 8, the resultant PCR products demonstrated the expected sizes of 608 (₁-AR), 194 (₂-AR), and 444 bp (₃-AR). The PCR product for -actin, an internal standard, was also detected in each preparation at its expected size of 387 bp.

View larger version (10K): [in this window] [in a new window]   Fig. 8.   Detection of 1-AR, 2-AR, and 3-AR mRNAs (run in triplicate) in the circular smooth muscle layer of the opossum IAS. Expected sizes of PCR products were 608, 194, 444, and 387 bp for 1-AR, 2-AR, 3-AR, and -actin, respectively. PCR products were electrophoresed in 2% agarose gel stained with ethidium bromide.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The studies demonstrate a systematic and comprehensive characterization of -adrenoceptors (-ARs) in the tonic smooth muscle of the gastrointestinal tract. The IAS smooth muscle served as the prototype using functional, classical pharmacology, molecular, and receptor binding approaches. The studies demonstrate: 1) the presence of membrane-bound -AR through Western blotting and -AR mRNA through RT-PCR; 2) the role of a heterogeneous population of -ARs (₁, ₂, and ₃) in mediating potent relaxation of the IAS smooth muscle; and 3) the presence of both high (₁/₂)- and low (₃)-affinity binding sites, with a significantly higher population of ₃-AR compared with ₁/₂.

The contribution of the three -AR subtypes in mediating IAS smooth muscle relaxation is in general agreement with previous reports in different smooth muscles including the GI tract (Goldberg and Frishman, 1995; De Ponti et al., 1996a; Roberts et al., 1997; Strosberg, 1997). The conclusions are based on the ability of ₁-, ₂-, and ₃-agonists to cause a full relaxation that is selectively antagonized by their respective antagonists. ZD 7114, a ₃-selective agonist (Growcott et al., 1993b), produces a concentration-dependent relaxation of the IAS smooth muscle that is antagonized by the ₃ antagonist SR 59230A (De Ponti et al., 1996b) but not by CGP 20712A or ICI 118551 (₁- and ₂-AR antagonists, respectively). The affinity values for antagonism by SR 59230A (pA₂ of 7.8) are consistent with previous studies in guinea pig ileum (pA₂, 7.7) (Roberts et al., 1999) and human colon (pA₂, 8.3) (De Ponti et al., 1996b). Procaterol, a ₂-selective agonist (Kotsonis and Majewski, 1994), also causes a concentration-dependent relaxation of the smooth muscle strips with a pEC₅₀ of 7.6, whereas ICI 118,551 (₂-selective antagonist) (Bilski et al., 1983) antagonizes this relaxation with a pA₂ value of 7.7. This is consistent with pA₂ values reported by Strosberg (1997). Likewise, xamoterol (₁ agonist) (Malta et al., 1985), causes a concentration-dependent relaxation of the IAS smooth muscle that is selectively antagonized by CGP 20712A (₁-AR antagonist).

In the rat (Roberts et al., 1999; Brown and Summers, 2001) and mouse (Hutchinson et al., 2001) ileum, it has been shown that ₃-ARs play a predominant role, whereas ₁-ARs have only a small role in smooth muscle relaxation. The presence of atypical or ₃-ARs was established in rat ileum by [¹²⁵I]CYP binding studies (Roberts et al., 1995) and by ₃-mRNA on RT-PCR analysis (Roberts et al., 1999). Roberts et al. (1995, 1999) were able to show the presence of ₃-AR but not those of ₁- and ₂-AR binding sites, even under classical binding conditions. Roberts et al. (1995, 1999) did, however, find an abundance of ₂-AR mRNA, in addition to ₃. Nevertheless, the role of ₂-AR was discounted because smooth muscle relaxation caused by zinterol (₂-AR-selective agonist) was antagonized by the ₃ antagonist SR 58894A and not by ICI 118551. The exact reason for the differences in the functional, binding, and molecular findings in these studies has not been fully delineated.

In contrast, the results of our studies in the IAS are in agreement with those in human colonic smooth muscle (De Ponti et al., 1996b) showing that the respective ₁- and ₂-selective antagonists CGP 20712A and ICI 118551 inhibit isoprenaline-mediated relaxation, which is further inhibited by SR 59230A. Differences between various studies may be reconciled on the basis of variations in species and tissues. The present studies, like those in human colon (De Ponti et al., 1996b), were conducted in spontaneously tonic smooth muscle, compared with others in which contraction was elicited by different contractile agonists. Whether such contractile agonists have attenuating effects in the functional expression of different -ARs remains to be determined.

Receptor binding, Western blot, and RT-PCR studies provide additional support in favor of the functional data. The receptor binding studies demonstrate, for the first time in the GI tract, the presence of two binding sites. These binding sites correspond to high affinity (R_H) ₁/₂ and low affinity (R_L) ₃ sites. We identified these binding sites with K_d values of 96 pM and 1.96 nM, respectively. The K_d values of the respective binding sites are similar to those described in adipocytes and Chinese hamster ovary (CHO) cells (Feve et al., 1991).

Two classes of binding sites were identified using competition studies with -AR subtype-selective antagonists. The rank order potency of the antagonists at the high-affinity site is ICI 118551 > CGP 20712A > SR 59230A with K_i values of 3.04 × 10⁸, 1.14 × 10⁷, and 8.53 × 10⁷ M, respectively. When radioligand concentrations were employed in the low-affinity range (1.61 nM), the potency was reversed, with SR 59230 > ICI 118551 > CGP 20712A. The corresponding K_i values with these antagonists were 4.81 × 10⁸, 6.80 × 10⁷, and 1.78 × 10⁶ M, respectively. The K_i values of CGP 201712A and ICI 118551 at the R_H are consistent with those reported at ₁- and ₂-ARs in CHO cells (Mejean et al., 1995). The K_i value for SR 59230A at the R_L is similar to that of the ₃-AR found in rat colon (Manara et al., 1995a). The K_i values of CGP 20712A and ICI 118551 are similar to those reported in guinea pig ileum and vascular smooth muscles (Kohout et al., 2001).

ZD 7114 was first described as a selective ₃-AR agonist in brown fat and guinea pig ileum (Holloway et al., 1991). Some subsequent studies have described ZD 7114 as having atypical ₃-AR antagonistic effects in certain tissues (Growcott et al., 1993a). In the IAS smooth muscle, ZD 7114 behaves as a full ₃-AR-selective agonist causing relaxation that is potently inhibited by SR 59230A. Therefore, the actions of ZD 7114 may be tissue- and species-specific.

SR 59230A was developed as the first ₃-AR-selective antagonist for the gut (Manara et al., 1995a). Recently, Horinouchi and Koike (2001) have raised the possibility that the effects of SR 59230A are tissue-specific. In the guinea pig gastric fundus and duodenum, SR 59230A may possess atypical -AR-agonistic activity by recognizing an aminotetralin moiety in the -AR. In our study, however, SR 59230A was found to be a selective ₃-AR antagonist with a pA₂ value of 7.8. It causes a concentration-dependent rightward shift in the CRC of ZD 7114 without modifying the effects of ₁-and ₂-AR agonists. In addition, SR 59230A alone does not cause a fall in IAS basal tone at concentrations up to 1 × 10⁴ M. It is possible that the presence of a bulky group on the arylethanolamine or aryloxypropanololamine side chain on both ZD 7114 and SR 59230A (Horinouchi and Koike, 2001) may render the receptor tissue- and species-specific. However, the opposing actions of ZD 7114 and SR 59230A in the IAS may not support that concept.

Receptor binding analysis reveals a higher receptor density of ₃-AR in the IAS smooth muscle. This is supported by the severalfold higher B_max in the case of low-affinity -AR (₃-AR) compared with high-affinity -AR (₁/₂-AR). With this information, one would have expected higher potencies of ₃- versus ₁- and ₂-AR agonists in causing IAS smooth muscle relaxation. The functional studies, however, show that in this respect, ₁-, ₂-, and ₃-agonists are nearly equipotent. We speculate three possible explanations for this disparity. The first and simplest explanation is the lack of effective ₃-AR agonists as compared with ₁- and ₂-AR agonists for the IAS smooth muscle at the present time. Second, ₃-AR in the IAS smooth muscle may have a large number of spare receptors. Third, ₃-AR may represent a heterogeneous population such as _3a- and _3b-ARs, as suggested by the recent studies in CHO (Hutchinson et al., 2002). Furthermore, the activation and signal transduction of such a _3a- and _3b-AR complex may prevent the full potency of the ₃-AR agonist. Therefore, it is no surprise that the ₃-AR agonist ZD 7114 has variable effects in different GI smooth muscle preparations (Growcott et al., 1993a,b). The involvement of _3a- and _3b-AR complex and the exact signal transduction involved in ₃-AR-mediated IAS relaxation by agonists such as ZD 7114 remains to be determined.

In addition to receptor binding studies, the presence of -AR in the IAS smooth muscle is further demonstrated by Western blot and RT-PCR studies. Western blot studies using primary antibodies specific to each -AR subtype reveal the presence of all three subtypes of -AR (₁-AR, 63 kDa; ₂-AR, 68 kDa; and ₃-AR, 65 kDa) in the rectum and IAS membranes. RT-PCR amplification was used to detect ₁-, ₂-, and ₃-AR in the circular smooth muscle layer of the IAS. The PCR products demonstrated the expected sizes of 608 (₁-AR), 194 (₂-AR), and 444 bp (₃-AR).

The present studies, therefore, provide comprehensive evidence for the presence and actions of ₁-, ₂-, and ₃-AR in IAS smooth muscle. In light of these findings, combined with the previously described actions of ₃-AR activation in the lower esophageal sphincter (DiMarino et al., 2002) with limited side effects and prolonged smooth muscle relaxation, we suggest that ₃-AR agonists in particular may have considerable physiological and therapeutic implications in anorectal and other spastic gastrointestinal motility disorders.