Introduction

Top Abstract Introduction Methods Results Discussion References

Smooth muscle from tissues such as the gastrointestinal tract (Michel and Whiting, 1988; Zhang et al., 1991; Gomez et al., 1992), uterus (Eglen et al., 1989), urinary bladder (Monferini et al., 1988; Noronha-blob et al., 1989), ciliary body (WoldeMussie et al., 1993) and airways (Gies et al., 1989; Eglen et al., 1994) of various species contains a mixture of M₂ and M₃ muscarinic receptors. The smooth muscle of the ileum also contains M₂ and M₃ receptors, as demonstrated in radioligand binding experiments (Giraldo et al., 1988; Michel and Whiting, 1990), mRNA analysis (Maeda et al., 1988) and immunoprecipitation studies with subtype-selective antibodies (Wall et al., 1991; Dörje et al., 1991). Under standard conditions the contractile response of the ileum is mediated by the M₃ muscarinic receptor, which couples to phospholipase C both in ileal smooth muscle (Candell et al., 1990) and in cells transfected with the M₃ receptor (Peralta et al., 1988). However, this subtype comprises only about 20% of the total muscarinic receptors in the ileum (Candell et al., 1990; Michel and Whiting, 1990). The M₂ muscarinic receptor accounts for the remaining 80% of the muscarinic receptor population and is known to mediate a pertussis toxin-sensitive inhibition of adenylyl cyclase in ileal smooth muscle (Candell et al., 1990), in the heart (Ehlert et al., 1989) and in cells transfected with the M₂ subtype (Kurose et al., 1983). In intact cell preparations of the longitudinal muscle of the guinea pig ileum, activation of M₂ receptors inhibits the accumulation of cAMP stimulated by beta adrenergic agonists and various other agonists (Reddy et al., 1995). In contrast, M₂ receptors inhibit only the increase in cAMP elicited by isoproterenol and forskolin in the longitudinal muscle of the rat ileum (Griffin and Ehlert, 1992). Because cAMP has been shown to relax smooth muscle (Berridge, 1975), it has been suggested that M₂ receptors may play a role in contraction by inhibiting the relaxant effects of agents that increase cAMP (Candell et al., 1990; Griffin and Ehlert, 1992).

This prediction has been borne out in studies on smooth muscle that had been treated with 4-DAMP mustard to inactivate most of the M₃ receptors. Such treatment causes a 20- to 40-fold reduction in the potency of highly efficacious agonists when contractions are measured under standard conditions in the ileum and trachea (Thomas et al., 1993; Watson et al., 1995; Thomas and Ehlert, 1996). The large decrease in agonist potency can be attributed to the loss of M₃ receptors. However, if contractions are measured in the same tissues in the presence of histamine and forskolin, a highly potent muscarinic response occurs that is blocked by selective antagonists in a manner consistent with an M₂ response (Thomas et al., 1993; Thomas and Ehlert, 1996). Presumably, the mechanism for this contraction involves an M₂-mediated inhibition of the relaxant effects of forskolin. This M₂-mediated contractile response is pertussis toxin-sensitive, unlike the standard contractile response, which is insensitive to pertussis toxin (Thomas and Ehlert, 1994, 1996). Using the same strategy, little or no contractile effects were detected for M₂ receptors in the rat fundus and guinea pig esophagus (Thomas and Ehlert, 1996). M₂ receptors have also been shown to inhibit the relaxant effects of isoproterenol on histamine-induced contractions of the ileum but not of the trachea (Thomas et al., 1993; Reddy et al., 1995; Watson et al., 1995). The reason for this difference between relaxant agents is unclear but may be related to the greater ability of forskolin to increase cAMP levels, compared with isoproterenol. Other investigators have used a different technique to detect a role for the M₂ receptor in contraction of the trachea (Fernandes et al., 1992; Watson and Eglen, 1994). However, see work by Roffel et al. (1993, 1995) for an opposing viewpoint.

It has long been known that isoproterenol is more effective at relaxing histamine-induced contractions, compared with those elicited by a muscarinic agonist (Van Amsterdam et al., 1989; Roffel et al., 1993). These observations have been interpreted as evidence that activation of M₂ receptors inhibits the cAMP accumulation elicited by isoproterenol, thereby reducing its relaxant effects against M₃-induced contractions, whereas histamine is without effect on cAMP levels. Alternatively, it has been proposed that cross-talk may occur between phospholipase C activation (via M₃ receptor stimulation) and the beta adrenergic signal transduction pathway (Roffel et al., 1993). A correlation between phosphoinositide hydrolysis and the shift in beta adrenergic receptor potency has been demonstrated (Van Amsterdam et al., 1989), and this effect may involve protein kinase C inactivation of G_s, resulting in functional uncoupling of beta adrenergic receptors (Pyne et al., 1992; Grandordy et al., 1994). Both M₂ and M₃ receptor-mediated antagonism of isoproterenol relaxant potency may be at work to varying degrees in different tissues, or a yet unknown response may be mediated by the M₂ receptor. Regardless, the functional role of the M₂ receptor in contraction has not been clearly defined in all smooth muscle tissues.

The purpose of the studies in the present report was to investigate the ability of the M₂ receptor in the guinea pig ileum to inhibit both the cAMP accumulation and the relaxant effects of a variety of agents known to stimulate adenylyl cyclase in other cells and tissues. Our results are consistent with the postulate that the functional role of the M₂ receptor in contraction depends on the amount of cAMP accumulation stimulated by the relaxant agent and the extent to which the M₂ receptor inhibits the increase in cAMP.

	Methods

Top Abstract Introduction Methods Results Discussion References

cAMP accumulation. Male Hartley guinea pigs (250-300 g) were sacrificed by asphyxiation with CO₂, followed by exanguination. Their ilea were immediately dissected, and the longitudinal muscle was removed by the method of Rang (1964) and immediately placed in ice-cold KRB buffer (124 mM NaCl, 5 mM KCl, 1.3 mM MgCl₂, 26 mM NaHCO₃, 1.2 mM KH₂PO₄, 1.8 mM CaCl₂, 10 mM glucose) gassed with O₂/CO₂ (19:1). Strips of muscle were cross-chopped at 350 µm with a McIlwain tissue chopper, washed extensively and equilibrated at 37°C for 15 min. In some experiments, tissues were incubated with the aziridinium ion of 4-DAMP mustard (40 nM) for 1 hr in the presence of AF-DX 116 (1 µM). After this incubation the tissue was washed extensively to remove 4-DAMP mustard and AF-DX 116. The slices were incubated in 10 ml of KRB buffer with [³H]adenine (1.0 µM, 50 µCi) for 40 min at 37°C, to allow uptake and incorporation into endogenous ATP. Slices were washed three times to remove extracellular [³H]adenine and were allowed to equilibrate for 15 min. Aliquots (70-100 µl) of gently packed tissue slices were incubated for 10 min at 37°C in KRB buffer (0.7 ml) containing 3-isobutyl-1-methylxanthine (0.5 mM) and various agonists. Ro2-01724 was substituted for 3-isobutyl-1-methylxanthine in experiments where 2-chloroadenosine or CGS-21680 was used, because of the ability of 3-isobutyl-1-methylxanthine to block adenosine receptors. Reactions were stopped by the addition of 0.3 ml of trichloroacetic acid (30%, w/v). Approximately 2000 cpm of [³²P]cAMP was added to each sample as an internal standard. The tubes were centrifuged, and the [³H]cAMP and [³H]ATP were separated from the supernatant fraction using the chromatography method described by Salomon et al. (1974). The supernatant was applied to a cation exchange column (1.25 ml of Dowex AG 50W-X4, 200-400 mesh) and washed twice with 1.25 ml of water. This eluate was collected, and the radioactivity was measured as the incorporation of [³H]adenine into [³H]ATP. The Dowex column was positioned over a column of neutral alumina (0.6 g), and the [³H]cAMP was eluted onto the alumina column with 5 ml of water. The [³H]cAMP was eluted from the alumina with 4 ml of 0.1 M imidazole HCl (pH 7.5). This fraction was collected and the radioactivity was measured to determine the amount of accumulated [³H]cAMP. The [³H]cAMP values were corrected for recovery of the internal standard and are expressed as a percentage of the amount of incorporation of [³H]adenine, to correct for any variation in the amount of tissue added to each assay.

Isolated ileum. Male Hartley guinea pigs were sacrificed as described above, and the whole ileum was rapidly removed. The most distal 10 cm of ileum was discarded and 2- to 3-cm ileal segments were cut, flushed with KRB buffer to remove ileal contents and mounted longitudinally in an organ bath containing KRB buffer at 37°C, gassed with O₂/CO₂ (19:1). Isometric contractions were measured with a force transducer and recorded on a polygraph and are expressed as the mass (grams) required to generate the measured force. The ileum was equilibrated for 1 hr at a resting tension equivalent to a load of 0.5 g. Three test doses of the muscarinic agonist oxo-M or histamine were added to the bath to ensure reproducibility of the preparation. Ileal segments that did not achieve >60% of the maximum from the test doses were discarded. Between each test dose the ileum was washed with fresh KRB buffer and incubated for 5 min. To calculate an EC₅₀ value for oxo-M, 6 to 10 concentrations of the compound, spaced geometrically every 0.33 log units, were added cumulatively to the bath, and contractile responses were recorded. After an EC₅₀ value for oxo-M was obtained, the ileum was washed and incubated for 30 min before additional measurements were made. In some experiments, tissues were incubated with the aziridinium ion of 4-DAMP mustard (40 nM) for 1 hr in the presence of AF-DX 116 (1 µM). Tissues were washed extensively to remove AF-DX 116 and unreacted 4-DAMP mustard. When an EC₅₀ value for oxo-M was to be obtained in the presence of AF-DX 116, the antagonist was incubated with the ileum for 30 min before contractions were measured.

Formation of the aziridinium ion of 4-DAMP mustard. 4-DAMP mustard undergoes two sequential reactions in aqueous solution at neutral pH. The first of these is the cyclization to its reactive aziridinium ion, and the second is the hydrolysis of the aziridinium ion to the stable alcohol product. In all experiments in which it was used, 4-DAMP mustard (10 µM) was first incubated in 10 mM phosphate buffer (pH 7.4) at 37°C for 30 min, to allow formation of the reactive aziridinium ion (Thomas et al., 1992). Immediately after cyclization, the solution of the aziridinium ion was placed on ice and used immediately.

Data analysis. Oxo-M inhibition of cAMP accumulation was calculated in two ways. Total inhibition (I_t) was calculated for each experiment with the following equation:

(1)

where O denotes the fold stimulation of cAMP in the presence of oxo-M and C denotes the fold stimulation of cAMP in the absence of oxo-M. The mean ± S.E.M. of the paired data are expressed in table 1. Inhibition of agonist-stimulated cAMP accumulation (I_a) was calculated with the following equation:

(2)

where C and O are defined as above (mean values used), b_o denotes the basal cAMP accumulation in the presence of oxo-M (mean of 0.65-fold) and b_c denotes the basal cAMP accumulation in the absence of oxo-M (1.0-fold). The EC₅₀ values of oxo-M (concentration of oxo-M required for half-maximal response) for contraction and inhibition of cAMP accumulation were estimated by nonlinear regression analysis of the data according to an increasing or decreasing logistic equation, as described previously (Candell et al., 1990). In some instances, a two-component model was used to analyze the contractile response data measured in 4-DAMP mustard-treated tissue in the presence of histamine and a relaxing agent. For this analysis, the following equation was fitted to the data by nonlinear regression analysis:

(3)

In this equation, Max denotes the maximal contractile response, Max_H denotes the proportion of the high-potency component, H1 and H2 denote the high- and low-potency Hill coefficients and EC_50H and EC_50L denote the EC₅₀ values of the high- and low-potency components of the contractile response curve. The contractile response data measured in the absence of AF-DX 116 were analyzed according to equation 1. The data measured in the presence of AF-DX 116 were also analyzed according to equation 1, except with the high- and low-potency EC₅₀ values multiplied by constants (shift_H and shift_L, respectively) corresponding to the factors by which AF-DX 116 increased the EC₅₀ values. In this analysis, the data obtained in the presence and absence of AF-DX 116 were analyzed simultaneously, sharing the estimates of Max, Max_H, H1, H2, EC_50H and EC_50L between the two sets of data. In most of the analyses, the estimates of shift_H and shift_L were set at constants corresponding to the expected values for AF-DX 116 (1.0 µM) at M₂ and M₃ receptors (see "Results").

Compounds. 4-DAMP mustard was synthesized in our laboratory as described previously (Thomas et al., 1992). Radiolabeled chemicals were obtained from ICN Biochemicals (Costa Mesa, CA). Cicaprost was obtained as a generous gift from Dr. Fiona McDonald of Schering AG (Berlin, Germany). AF-DX 116 was acquired from Boehringer Ingelheim Pharmaceuticals (Ridgefield, CT). SKF-38393, CGS-21680 and oxo-M were obtained from Research Biochemicals Inc. (Natick, MA). All other drugs and chemicals were obtained from Sigma Chemical Co. (St. Louis, MO).

	Results

Top Abstract Introduction Methods Results Discussion References

Muscarinic inhibition of agonist-stimulated cAMP accumulation. We measured the ability of the highly efficacious muscarinic agonist oxo-M to inhibit the cAMP accumulation elicited by a variety of agents that have been shown to stimulate adenylyl cyclase in other tissues. The cAMP-stimulating agents were used at concentrations from 0.8 to 10 µM, whereas oxo-M was always used at a nearly maximally effective concentration of 1 µM. The largest increase in cAMP was elicited by forskolin (10 µM), which stimulated cAMP levels 14.9-fold, from a basal value of 0.35 ± 0.04% (expressed as a percentage of the total [³H]adenine-labeled nucleotides) to 5.35 ± 1.32%. Moderate stimulation of cAMP (2.51-1.52-fold stimulation over basal) was observed with isoproterenol (1 µM), PGE₁ (10 µM), PGE₂ (10 µM), cicaprost (0.8 µM) and PGI₂ (10 µM), with stimulations over basal being 2.51-, 2.27-, 2.28-, 2.45- and 1.52-fold, respectively (table 1). Little or no cAMP stimulation was observed with dopamine, 5-HT, 5-MT, dimaprit, VIP, SKF-38393, 2-chloroadenosine, CGS-21680, prostaglandin D₂, secretin and vasopressin. Oxo-M (1 µM) inhibited basal cAMP levels by 35%. Indomethacin and tetrodotoxin had no effect on the oxo-M inhibition of basal cAMP (data not shown). Oxo-M (1 µM) inhibited the cAMP measured in the presence of the various agents by 21 to 67% (total inhibition). PGI₂ and cicaprost were exceptions, with oxo-M exerting total inhibition of only 12 and 17%, respectively. Because part of the total inhibition can be attributed to inhibition of basal cAMP levels, we also tabulated the percent inhibition of agonist-stimulated cAMP, as described in "Methods." This percentage was calculated after the basal values in the absence (1.0-fold) and presence (0.65-fold) of oxo-M were subtracted from the agonist-stimulated values in the absence and presence of oxo-M, respectively (table 1). Such a calculation estimates the specific inhibition of each agonist effect by muscarinic receptor activation. Using this calculation, oxo-M specifically inhibited forskolin-, isoproterenol-, PGE₁-, PGE₂-, dopamine- and VIP-stimulated cAMP levels by 70, 38, 20, 29, 26 and 39%, respectively. PGI₂-, cicaprost-, 5-MT- and dimaprit-stimulated cAMP levels were not specifically inhibited by oxo-M (0, 5, 0 and 0%, respectively).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Oxo-M inhibition of cAMP accumulation stimulated by 10 µM forskolin (), 1 µM isoproterenol (), 0.8 µM cicaprost () and 10 µM dopamine (). Data are expressed as percent inhibition of stimulated cAMP accumulation (as percent conversion), without subtraction of basal levels. All tissue was treated with 4-DAMP mustard as described in "Methods." Each point represents the mean ± S.E.M. of five experiments.

Maximal relaxant effects of heterologous agonists. We investigated the ability of each agonist to inhibit histamine-induced contractions in the guinea pig ileum, to identify those agents having a correlation between their stimulation of cAMP accumulation and their relaxant ability. The isolated ileum was contracted with histamine (0.3 µM) and allowed to stabilize, and then a maximal concentration of the cAMP-stimulating agent was added. The amount of relaxation recorded is expressed as a percentage of the original histamine contraction (table 3). For the purpose of making the comparison, the agonists are listed in order of decreasing effectiveness in stimulating cAMP accumulation. In the rightmost column, the percentage relaxation value for each agonist is listed together with its rank order (in parentheses). The relaxant effects of forskolin, isoproterenol and cicaprost were rather large, representing 103, 97.5 and 43.9% inhibition, respectively, of the histamine-induced contraction. Dopamine, VIP, CGS-21680, 2-chloroadenosine and SKF-38393 caused a small amount of relaxation (13-30%).

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Comparison of cAMP accumulation (fold stimulation over basal level) and relaxant activity (percent relaxation of histamine-induced contraction) of maximal concentrations of various receptor agonists. Each data point represents a mean of 3 to 10 experiments.

Relaxant effects of agonists on muscarinic and histamine-induced contractions. The potencies of the various agents for inhibiting histamine- and oxo-M-mediated contractions were compared. Only those agents that demonstrated significant relaxant effects (forskolin, isoproterenol, dopamine and cicaprost) were investigated. For these experiments, the isolated ileum was precontracted with either histamine (0.3 µM) or oxo-M (40 nM). These concentrations were chosen because they reliably produced contractions of equal size. Over the course of these studies, the estimates of the average tension ± S.E.M. (expressed in units of mass) elicited by histamine and oxo-M were 3.20 ± 0.22 g and 3.33 ± 0.20 g, respectively (not statistically different, P = .67). Cumulative concentration-relaxation curves were measured for each relaxing agent against histamine and oxo-M (fig. 3). The EC₅₀ value and maximal relaxant effect for each agent are listed in table 4. In general, each agent was more effective at inhibiting histamine-induced contractions, compared with oxo-M-induced contractions. At high concentrations, forskolin caused complete inhibition of both histamine- and oxo-M-induced contractions. In contrast, the maximal relaxant effect of the other agents was always much less when oxo-M was used as the contractile agent, compared with histamine.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Concentration-response curves for relaxation by various agents (A-D) against contractions induced by 0.3 µM histamine () and 40 nM oxo-M (). Data are presented as the percent decrease in tension of the initial contraction. Each point represents the mean ± S.E.M. of three experiments.

Contractile responses in 4-DAMP mustard-treated ilea. To determine whether the M₂ receptor could elicit an indirect contraction by preventing the relaxant effects of the various cAMP-stimulating agents, we measured contractile responses to oxo-M in isolated ileum according to the experimental design shown in figure 4. In the first phase of the experiments (treatment phase), the ileum is incubated with 4-DAMP mustard (40 nM) in the presence of AF-DX 116 (1 µM) for 1 hr and then washed extensively (Thomas et al., 1993). This treatment inactivates most of the M₃ receptors without affecting the M₂ receptors. In the second phase (test phase), the ileum is contracted with a nearly maximally effective concentration of histamine, and then a relaxant agent is added. After a stable relaxation is achieved, increasing concentrations of oxo-M are added to the bath so that cumulative concentration-response curves can be measured. These test phase measurements are repeated in the presence of AF-DX 116 to characterize the muscarinic receptor subtypes mediating the contraction. Experiments carried out under these conditions are shown in figure 5, B to E. In figure 5, the horizontal lines denote the size of the contraction elicited by histamine (0.3 µM) alone (upper line) and in combination with the relaxant agent (lower line). A summary of EC₅₀ values and Hill coefficients is found in table 5.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Schematic pen tracing illustrating the experimental protocol of the treatment and test phases when contractile responses are measured in ilea treated with 4-DAMP mustard.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effects of AF-DX 116 on the contractile response to oxo-M in tissues treated with 4-DAMP mustard under standard conditions (A) and conditions of elevated cAMP (B-E). Except under standard conditions (A), ileal segments were contracted with histamine (0.3 µM) and relaxed with 10 µM forskolin (B), 1 µM isoproterenol (C), 10 µM dopamine (D) or 0.8 µM cicaprost (E) before oxo-M-induced contractions were measured. , Control; , 1 µM AF-DX 116. In all experiments, tissues were treated with 4-DAMP mustard as described in "Methods." Each point represents the mean ± S.E.M. of four to eight experiments.

(4)

	Discussion

Top Abstract Introduction Methods Results Discussion References

Reddy et al. (1995) have demonstrated that a variety of agents, including beta adrenergic agonists and prostaglandins, stimulate cAMP accumulation in slices of the longitudinal muscle of the guinea pig ileum. Those investigators measured the cAMP response to each agonist at a concentration of 10 µM and in the presence of a low concentration of forskolin (0.1 µM), which enhanced the cAMP response. Among the compounds tested, PGE₁ elicited the largest increase in cAMP over basal levels (5.3-fold). Intermediate increases in cAMP were noted with PGE₂ (3.4-fold) and isoproterenol (3.0-fold), whereas the smallest increases were observed with VIP (2.3-fold), 5-HT (1.7-fold) and the beta-3-selective agonist BRL 37344 (1.4-fold). In the present study, we observed a similar pattern of cAMP stimulation with these agonists, although the absolute stimulation was about 50% less, on average. This decrease can be attributed to the lack of forskolin (0.1 µM) in our assays, because Reddy et al. reported that forskolin enhanced the cAMP response to isoproterenol (10 µM) by 40%. We also observed appreciable stimulation of cAMP by the prostanoid derivative cicaprost (2.45-fold), PGI₂ (1.52-fold), the H₂ histamine agonist dimaprit (1.32-fold) and 5-MT (1.25-fold).

When measured in the presence of the highly efficacious muscarinic agonist oxo-M (1.0 µM), the cAMP accumulation elicited by the various agonists was substantially inhibited. Presumably, this muscarinic inhibition is mediated by the M₂ receptor because, in both rat and guinea pig ileum, subtype-selective antagonists block the inhibition in a manner consistent with an M₂ response (Candell et al., 1990; Griffin and Ehlert, 1992; Reddy et al., 1995). In the present study, both basal and agonist-stimulated cAMP accumulation was inhibited by oxo-M. To estimate that component of the muscarinic inhibition that could be attributed to specific inhibition of agonist-stimulated cAMP accumulation, we subtracted the appropriate basal values from the data, as described in "Methods," before estimating the percentage inhibition. These calculations yielded an estimate of 70% for the inhibition of forskolin-stimulated cAMP accumulation by oxo-M. In contrast, oxo-M inhibited the cAMP accumulation elicited by isoproterenol, PGE₁, PGE₂, dopamine, 5-HT and VIP to a lesser extent (38, 20, 29, 26, 23 and 39%, respectively) and was without effect on the cAMP accumulation elicited by cicaprost, PGI₂, 5-MT or dimaprit. Thus, activation of M₂ muscarinic receptors inhibits the increase in cAMP elicited by almost all of the agonists tested. In the rat ileum, M₂ receptors mediate inhibition of forskolin- and isoproterenol-stimulated cAMP accumulation but not that elicited by PGE₁ or PGE₂ (Griffin and Ehlert, 1992). The different responses of these two tissues may be due to the greater abundance of M₂ receptors in the guinea pig ileum (Michel and Whiting, 1990; Candell et al., 1990).

In addition to cAMP, other mediators elicit relaxation in smooth muscle, including cyclic GMP, which may be stimulated either directly or indirectly (through nitric oxide synthase pathways). To identify agents that cause relaxation through cAMP, we measured the ability of the agents listed in table 1 to inhibit histamine-induced contractions, so that it would be possible to identify agents showing a positive correlation between cAMP accumulation and relaxation. Among the compounds tested, only forskolin, isoproterenol, cicaprost and dopamine caused appreciable relaxation. CGS-21680, 2-chloroadenosine, SKF-38393 and VIP elicited small inconsistent relaxations and other agents, such as PGE₁, PGE₂, PGI₂, 5-HT, 5-MT and dimaprit, caused small contractions, presumably due to their respective abilities to activate multiple receptor subtypes, including some coupled to phosphoinositide hydrolysis. Among the relaxant agonists, there was a general correlation between the potential of agonists to stimulate cAMP and their ability to cause relaxation, suggesting that cAMP is the predominant mechanism for relaxation (fig. 2). One notable exception was cicaprost, the prostanoid IP receptor-selective agonist, which caused proportionately greater cAMP accumulation than relaxation. Lawrence et al. (1992) characterized prostanoid receptors in the guinea pig ileum and found that the IP receptor is implicated in direct smooth muscle relaxation and, to a lesser extent, neuronal stimulation of acetylcholine release. Indeed, in our experiments with cicaprost, a small contractile component frequently preceded relaxation. This effect accounts for the fact that cicaprost caused less relaxation than predicted from its potential to stimulate cAMP accumulation.

It is well established that isoproterenol is more effective at inhibiting histamine-induced contractions in the trachea, compared with those elicited by muscarinic agonists (Van Amsterdam et al., 1989; Roffel et al., 1993). To explain the differential effects of isoproterenol, Torphy et al. (1983) suggested that muscarinic inhibition of adenylyl cyclase protects the muscarinic contractile mechanism from the cAMP-mediated relaxant effects of isoproterenol. It is now known, of course, that inhibition of adenylyl cyclase and stimulation of contractions are mediated by two different muscarinic receptors (i.e., M₂ and M₃ receptors, respectively) (Peralta et al., 1988; Candell et al., 1990). Histamine lacks this two-pronged mechanism, so its contractile response is more sensitive to the relaxant effects of isoproterenol. Although direct evidence for a role of the M₂ receptor in inhibiting the relaxant effects of isoproterenol on M₃-mediated contractions in the trachea and ileum has been obtained by several investigators (Thomas et al., 1993; Watson et al., 1995), it is unclear whether the M₂ mechanism is the only mechanism that accounts for the differential relaxant effects of isoproterenol (Roffel et al., 1995). We chose to compare the relaxant potential of the various cAMP-stimulating agents against histamine- and oxo-M-induced contractions, to obtain a simple measure of whether the M₂ receptor diminished the relaxant effect of the agent. If an agent was more effective at inhibiting histamine-induced contractions, compared with those elicited by oxo-M, then a role for the M₂ receptor might be inferred. The greatest differential relaxant effects were noted with forskolin, isoproterenol and dopamine, whereas little or no discrimination was noted with cicaprost (fig. 3). Oxo-M inhibition of cAMP levels stimulated by each of these agents (table 1) followed a similar pattern. Thus, there appears to be at least a rough correlation between the extent to which the M₂ receptor inhibits the cAMP accumulation elicited by a drug and the extent to which the drug displays differential relaxant effects.

In examining the data shown in figure 3, it becomes apparent that there is no simple way to quantify the differential relaxant effects of the various agents. Forskolin exhibited a 6-fold difference in relaxant potency between histamine and oxo-M but no change in maximal response, whereas the other agents showed differences in both potency and maximal response. If the mechanism for the reduced relaxant activity in the presence of oxo-M is caused by M₂-mediated inhibition of cAMP accumulation, then, in the presence of oxo-M, relaxing agents should behave as if they are less efficacious at stimulating cAMP accumulation, and hence relaxation. A useful measure of the apparent decrease in intrinsic efficacy is the degree of receptor inactivation that would cause an equivalent loss of activity. We have used the term "stimulus inactivation" to denote this decrement in relaxant activity against oxo-M-induced contractions, and we have described a means of calculating this parameter in the "Appendix." Table 4 lists our estimates of this parameter. It can seen that there is a rough correlation between the ability of oxo-M to inhibit both the relaxant stimulus and the cAMP accumulation elicited by the various agents. Interestingly, this correlation becomes closer if the inhibition of cAMP accumulation is calculated by subtracting the control basal value (1.0-fold) from both the stimulated and inhibited values before calculating the percentage inhibition by oxo-M. When calculated in this manner, the oxo-M inhibition of the cAMP accumulation stimulated by forskolin (73%), isoproterenol (61%), dopamine (110%) and cicaprost (29%) is in agreement with the degree to which oxo-M inhibits the relaxant stimulus generated by the agents (82, 68, 93 and 4%, respectively).

As described above, forskolin was able to cause complete inhibition (100%) of both histamine- and oxo-M-induced contractions, whereas isoproterenol, dopamine and cicaprost showed reduced maximal responses against oxo-M, relative to histamine. This behavior can be rationalized with the relationship between relaxation and cAMP levels shown in figure 2. It can be seen that complete relaxation (100%) of histamine-induced contractions occurs when cAMP levels are stimulated approximately 2.5-fold over basal. Even though oxo-M inhibited forskolin-stimulated cAMP levels by 70%, the residual level of cAMP in the presence of forskolin and oxo-M (4.5-fold over basal) (table 1) still exceeded that required for maximal relaxation. In contrast, the cAMP levels elicited by isoproterenol, dopamine and cicaprost in the presence of oxo-M (1.59-, 0.96- and 2.03-fold, respectively) (table 1) were insufficient to cause complete relaxation.

To further explore the role of the M₂ receptor in opposing smooth muscle relaxation, we used the novel method outlined in figure 4, which was previously developed in our laboratory (Thomas et al., 1993). We first inactivated M₃ receptors with the irreversible alkylating agent 4-DAMP mustard and then measured responses to a muscarinic agonist in the presence of histamine and a relaxing agent. When forskolin was used as the relaxant agent, the K_B values of AF-DX 116, methoctramine, p-fluorohexahydrosiladifenidol and pirenzepine were in good agreement with their respective binding affinities for native and cloned M₂ receptors but much different from those reported for the M₁, M₃, M₄ and M₅ subtypes (Ehlert and Thomas, 1995; Esqueda et al., 1996). Thus, under the conditions of the experiment, oxo-M acts through the M₂ receptor to release the brake that forskolin has put on histamine-induced contractions. Thomas et al. (1993) and Reddy et al. (1995) have used this strategy to show that the M₂ receptor also opposes the relaxant effect of isoproterenol on histamine-induced contractions in the guinea pig ileum. In this study, we have reexamined forskolin and isoproterenol, so that data on contractions and cAMP accumulation could be compared for a group of relaxant agents.

When forskolin was used as the relaxing agent under the experimental conditions shown in figure 4, AF-DX 116 (1.0 µM) shifted the concentration-response curve for oxo-M to the right 25-fold, in a manner consistent with simple competition at an M₂ muscarinic receptor. The calculated K_B value (7.38) was nearly the same as the estimate of the binding affinity of AF-DX 116 for the cloned M₂ receptor (7.27) when measured in modified KRB buffer (Esqueda et al., 1996). The maximal response of oxo-M was only about 55% of that of the histamine-induced contraction, suggesting that the M₂ receptor cannot fully oppose the inhibitory effect of forskolin on histamine-induced contractions. When isoproterenol was used as the relaxant agent during the test phase, the concentration-response curve of oxo-M had a low slope and appeared biphasic, suggesting high- and low-potency components. AF-DX 116 simplified the curve by causing a greater apparent competition at the high-potency component. We reasoned that at low concentrations oxo-M was acting through the AF-DX 116-sensitive M₂ receptor to disinhibit histamine-induced contractions, whereas at high concentrations oxo-M might be eliciting contractions through M₃ receptors not inactivated by 4-DAMP mustard. The contribution of the M₃ receptor could account for the large maximal response, which frequently exceeded the histamine-induced contraction, although these did not differ statistically. Consequently, we analyzed the data according to a two-component model, assuming that AF-DX 116 (1.0 µM) should cause shifts of 20- and 2.3-fold in the M₂ and M₃ components of the concentration-response curve, respectively. Regression analysis yielded estimates of the size of the M₂ component when the different relaxing agents were used. The size of the M₂ component of the contractile response appears to be related to the amount of the relaxation induced by the specific relaxant agent and the extent to which the M₂ receptor inhibited the increase in cAMP elicited by the agent. When expressed in mass equivalents, the maximal contractile responses of the M₂ components were 1.8 and 1.6 g when forskolin and isoproterenol were used as relaxing agents, respectively. No M₂ component was apparent when cicaprost was used as the relaxant agent, in agreement with the one-site analysis, where the shift induced by AF-DX 116 was in close agreement with an M₃ response. These estimates correlate roughly with the extent to which oxo-M inhibited the cAMP accumulation elicited by forskolin (70%), isoproterenol (38%) and cicaprost (5%). Although oxo-M caused a 26% inhibition of dopamine-stimulated cAMP accumulation, dopamine elicited a relatively small relaxation, so that the M₂ component of the response was only 0.7 g. Collectively, these data show general agreement between the extent to which activation of the M₂ receptor opposes the ability of a drug to stimulate cAMP accumulation and its ability to inhibit histamine-induced contractions.

When forskolin was used as the relaxing agent during the test phase, the concentration-response curve of oxo-M was relatively simple and lacked the low-potency M₃ component characteristic of the data collected with the other relaxant agents. It seems likely that the higher levels of cAMP elicited by forskolin may have completely suppressed M₃-mediated contractions already greatly inhibited by 4-DAMP mustard.

The potency of oxo-M for eliciting contractions through the M₂ mechanism is very high. More specifically, when measured during the test phase in the presence of different relaxant agents, the EC₅₀ values of the M₂ component were in the range of 8 to 16 nM. In contrast, the EC₅₀ value of oxo-M for eliciting contractions through the M₃ receptor under standard conditions was approximately 30 to 40 nM. Moreover, after 4-DAMP mustard treatment, the EC₅₀ value of the standard contractile response was increased to 207 nM. Thus, even after the M₃ response has been greatly suppressed, oxo-M can elicit contractions through the M₂ mechanism with about 20-fold greater potency, underscoring the significance of this phenomenon.

When measured under the conditions of the test phase, M₂-mediated contractions were most easily detected when forskolin was used as the relaxing agent. Under these conditions, the contractile response to oxo-M yielded a relatively simple concentration-response curve that was shifted to the right 25-fold in the presence of the M₂-selective antagonist AF-DX 116 (1.0 µM). When the other relaxant agents were used, the M₃ receptor contributed to the contractile response, thereby diminishing the relative contribution of the M₂ receptor and making it more difficult to detect the M₂ component of the response. These observations correlate generally with our measurements of cAMP. Forskolin caused by far the largest increase in cAMP accumulation, and oxo-M inhibited forskolin-stimulated cAMP levels by 70%. In contrast, oxo-M inhibited isoproterenol-stimulated cAMP accumulation by only 38% and the cAMP levels elicited by the other agents by less. We previously suggested (Thomas and Ehlert, 1996) that these observations might explain, in part, why it was possible to measure an M₂-mediated inhibition of the relaxant effect of forskolin, but not isoproterenol, on histamine-induced contractions of the trachea (Watson et al., 1995). Detecting a role for the M₂ receptor in inhibiting the relaxant effects of isoproterenol in the trachea may be easier when more M₃ receptors are selectively inactivated. Regardless, M₂ receptors have been shown to inhibit the relaxant effects of forskolin on M₃-mediated contractions of the trachea (Thomas and Ehlert, 1996). Differences in the contractile effects of the M₂ receptor in different tissues might also be explained by a variation in the density and coupling efficiency of the receptor. A decrease in the latter might account for our inability to measure an M₂-mediated inhibition of the relaxant effect of forskolin on histamine-induced contractions of the rat fundus and guinea pig esophagus (Thomas and Ehlert, 1996).

In summary, our results confirm that the M₂ receptor has the role of inhibiting the relaxant effects of agents that increase cAMP in the guinea pig ileum. The involvement of the M₂ receptor in this role depends upon the amount of cAMP elicited by the agent and the extent to which the M₂ receptor opposes the increase in cAMP.