Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Muscarinic M₂ and M₃ receptors are abundantly expressed postjunctionally in smooth muscle, where they mediate the contractile effects of acetylcholine (see reviews by Eglen et al., 1996; and Ehlert et al., 1997). When measured in the absence of other heterologous agents, the contractile response to muscarinic agonists is inhibited by subtype-selective antagonists in a manner consistent with an M₃ mechanism. This behavior is consistent with the known coupling of M₃ receptors to pertussis toxin-insensitive G_q, which mediates phosphoinositide hydrolysis and mobilization of Ca²⁺ (Peralta et al., 1988; Noronha-Blob et al., 1989; Candell et al., 1990; Roffel et al., 1990). The muscarinic M₂ receptor is known to signal through G_i in smooth muscle to mediate pertussis toxin-sensitive responses including the inhibition of both adenylyl cyclase (Noronha-Blob et al., 1989; Candell et al., 1990; Yang et al., 1991) and Ca²⁺-activated potassium channels (Kotlikoff et al., 1992) as well as the stimulation of a nonselective cation conductance (Inoue and Isenberg, 1990; Bolton and Zholos, 1997). Ultimately, M₂-mediated effects on contraction through these latter mechanisms are conditional upon Ca²⁺ mobilization via another receptor. Consequently, the pharmacological paradigms required to demonstrate a contractile role of the M₂ receptor are more complicated than the standard assay used to establish a role for the M₃ receptor. Nevertheless, pharmacological studies employing pertussis toxin, irreversible M₃-selective muscarinic antagonists, and heterologous contractile and relaxant agents have demonstrated two roles for the M₂ receptor in contraction: an inhibition of the relaxation caused by agents that increase cAMP and a conditional potentiation of the M₃ receptor-mediated contractions (Thomas et al., 1993; Thomas and Ehlert, 1994, 1996; Hegde et al., 1997; Sawyer and Ehlert, 1998, 1999b; Shen and Mitchelson, 1998).

Recent studies on M₂ and M₃ receptor knockout mice are generally consistent with the results of pharmacological experiments on wild-type animals of other species showing that it is primarily the M₃ receptor that mediates a direct contraction in smooth muscle. In M₃ receptor knockout mice, the muscarinic contractile response of the ileum and urinary bladder are greatly reduced compared with wild-type mice (Matsui et al., 2000; Stengel et al., 2002), whereas a much smaller decrement in contractile function was noted in M₂ receptor knockout mice (Stengel et al., 2000). In mutant mice lacking both M₂ and M₃ receptors, the potent contractile response to muscarinic agonists in ileum and urinary bladder is completely eliminated, demonstrating that only M₂ and M₃ muscarinic receptors contribute to the direct contractile response in these tissues (Matsui et al., 2002). So far, no studies have been reported on the role of M₂ receptors in mediating an inhibition of the relaxant effects of agents that increase cAMP in muscarinic receptor knockout mice.

In this report, we describe the relaxant effects of isoproterenol and forskolin on muscarinic agonist-induced contraction of smooth muscle in wild-type and M₂ receptor knockout mice. Our results show that the relaxant effects of forskolin on muscarinic agonist-induced contractions, but not those elicited by KCl, are greatly increased in the ileum, bladder, and trachea of M₂ receptor knockout mice as compared with wild-type mice. Under similar conditions, the relaxant effect of isoproterenol was also enhanced in the ileum and urinary bladder, but not in the trachea. These results demonstrate that a component of the contractile mechanism of muscarinic agonists in smooth muscle involves an M₂ receptor-mediated inhibition of the relaxant effects of agents that increase cAMP levels.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. The generation of M₂ muscarinic receptor knockout mice has been described previously (Matsui et al., 2002). A mutant mouse line was established in a mixed background between 129/SvJ and C57BL/6. This line was backcrossed with C57BL/6 mice to yield an N3 generation of M₂/ muscarinic receptor knockout mice, which were used in the pharmacological studies described in this report. C57BL/6 mice were purchased from Harlan (Indianapolis, IN).

Isolated Smooth Muscle. Wild-type (M₂+/+) male C57BL/6 mice and male M₂ muscarinic receptor knockout (M₂/) mice were used in these studies. Their body weights were approximately 25 to 30 g. The mice were euthanized by CO₂ asphyxiation, and various smooth muscle preparations were immediately removed. Segments of whole ileum (1.5-2 cm in length) were excised starting at a point approximately 5 cm rostral from the ileocecal junction and mounted longitudinally in an organ bath with silk thread. The whole trachea was dissected free of adhering tissue and mounted as a single ring on stainless steel supports. The whole urinary bladder was removed and mounted longitudinally with silk thread. All tissues were bathed at 37°C within 50-ml organ baths containing Krebs-Ringer bicarbonate (KRB) buffer (124 mM NaCl, 5.0 mM KCl, 1.3 mM MgSO₄, 26 mM NaHCO₃; 1.2 mM KH₂PO₄, 1.8 mM CaCl₂, 10 mM glucose) gassed with O₂/CO₂ (19:1). Indomethacin (1.0 µM) was present in the KRB buffer at all times. Indomethacin is often included in experiments on tracheal smooth muscle to remove the influence of endogenous prostaglandins (Muccitelli et al., 1987), to dissipate spontaneous tracheal tone (Small et al., 1990), and to maintain receptor-stimulated tone (Mansour and Daniel, 1986), although the reasons for these uses are incompletely understood. To maintain consistency, all tissues were exposed to indomethacin. The tissues were connected to force-displacement transducers and isometric tension was recorded using either Polygraph (Grass Instruments, Quincy, MA) or PowerLab (ADInstruments, Grand Junction, CO) recording systems. Resting tensions were adjusted to loads equivalent to those generated by masses of 0.3, 2, and 0.5 g in the ileum, trachea, and urinary bladder, respectively. The tissues were first allowed to equilibrate for at least 60 min, and then three test doses of KCl (50 mM) were applied. After each test dose, the tissues were washed with fresh KRB buffer and allowed to rest for approximately 5 to 10 min. The contractile response to the third test dose of KCl was used in calculations to normalize the response to the muscarinic agonist, oxotremorine-M, relative to that elicited by KCl. The tissues were allowed to rest for 15 min, and then a cumulative concentration-response curve to oxotremorine-M was measured, with the agonist concentrations being spaced 3-fold. Only the tonic phase of contraction was used in the calculation of concentration-response curves. The tissues were washed with fresh KRB buffer and allowed to rest for 30 min. A second control concentration-response curve to oxotremorine-M was measured, and this second curve was used as the control to which the curves measured in the presence of relaxant agents were compared. Tissues were washed extensively with fresh KRB buffer after measurement of each concentration-response curve and were allowed to rest for 30 min before the next concentration-response curve was measured.

Calculations. The EC₅₀ value (concentration of agonist eliciting half-maximal contraction) and the maximal response (E_max) to oxotremorine-M were estimated from the concentration-response data by nonlinear regression analysis using an increasing logistic equation as described previously (Candell et al., 1990). In most cases, the relaxant agents caused a decrease in both the maximal response and potency of oxotremorine-M for eliciting contraction. In such instances, the effect of the relaxant agent could be simulated quantitatively by assuming that the relaxant agent caused a decrease in the proportion of receptors or the intrinsic efficacy of the receptor-oxotremorine-M complex. We refer to this phenomenon as the decrease in the "observed coupling efficiency" of oxotremorine-M caused by the relaxant agent. This measure of relaxant action was calculated as described previously (Ostrom and Ehlert, 1997) using a method akin to Furchgott analysis (Furchgott, 1966). This estimate is simply an empirical measure of the relaxant effect, and no conclusion about the mechanism of relaxation is deduced with this estimate. We simply use the estimate of observed coupling efficiency as a means of summating the inhibitory effects of the relaxant agent on the two disparate parameters, EC₅₀ and E_max. For each experiment, the estimates of EC₅₀ were converted to negative logarithms (pEC₅₀), and the effect of the relaxant agent on the EC₅₀ value was calculated as the difference between the control pEC₅₀ value and that measured in the presence of the relaxant agent (i.e., log EC₅₀ shift). To assess whether the relaxant agent had a significant effect on the EC₅₀ value, a t distribution was used to determine whether the log EC₅₀ shift values were significantly different from zero. Similarly, a t distribution was used to determine whether the logarithm of the observed coupling efficiencies were significantly different from a value of zero (i.e., no change in coupling efficiency). A paired t test was used to determine whether the relaxant agent had a significant effect on the E_max. To determine whether these estimates of relaxant activity in M₂ receptor knockout mice were significantly different from those measured in wild-type animals, an unpaired t test was used.

Drugs and Chemicals. The reagents used in this study were obtained from the following sources: oxotremorine-M, Sigma/RBI, Natick, MA; isoproterenol, indomethacin and tetrodotoxin, Sigma-Aldrich, St. Louis, MO; and forskolin, Calbiochem, San Dieog, CA.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Oxotremorine-M elicited contractions in ileum, trachea, and urinary bladder of wild-type mice with mean pEC₅₀ values ± S.E.M. of 6.70 ± 0.037, 6.94 ± 0.036, and 6.58 ± 0.044; and E_max values, expressed as mass equivalents, of 1.36 ± 0.083, 2.60 ± 0.23, and 4.3 ± 0.47 g, respectively. When expressed relative to the contraction elicited by KCl (50 mM), the E_max values of oxotremorine-M in these tissues were 188 ± 20, 218 ± 14, and 223 ± 11%, respectively. In M₂ receptor knockout mice, the potency of oxotremorine-M was lower. This decrease in potency corresponded to mean EC₅₀ values that were 2.1-, 1.21-, and 1.48-fold greater in the ileum, trachea, and urinary bladder, respectively. These differences reached statistical significance in ileum (P = 0.00013), but not in trachea (P = 0.124) or urinary bladder (P = 0.114). There were no significant differences between wild-type and M₂ receptor knockout mice with regard to the E_max values of oxotremorine-M expressed relative to the contraction elicited to KCl in the three tissues (P = 0.93, 0.12, and 0.11 in ileum, trachea, and urinary bladder, respectively). A similar conclusion was reached when E_max was expressed in units of mass equivalents. These data are summarized in Figs. 1 and 2 and Table 1. In a limited number of experiments, we found that tetrodotoxin (1 µM) was without effect on the contractile response to oxotremorine-M in wild-type and M₂ receptor knockout mice.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Effects of forskolin on oxotremorine-M-mediated contractions in ileum (a and d), trachea (b and e), and urinary bladder (c and f) from wild-type (a-c) and M2 muscarinic receptor knockout mice (d-f). The contractile response to oxotremorine-M was measured in the absence () and presence () of forskolin (5.0 µM). The magnitudes of the contractions are expressed relative to the contraction elicited by KCl (50 mM). The data represent the mean values ± S.E.M. from five wild-type and five knockout mice.

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Effects of isoproterenol on oxotremorine-M-mediated contractions in ileum (a and d), trachea (b and e), and urinary bladder (c and f) from wild-type (a-c) and M2 muscarinic receptor knockout mice (d-f). The contractile response to oxotremorine-M was measured in the absence () and presence () of isoproterenol (1.0 µM). The magnitudes of the contractions are expressed relative to the contraction elicited by KCl (50 mM). The data represent the mean values ± S.E.M. from five wild-type and five knockout mice.

Figure 1 shows the relaxant effect of forskolin (5.0 µM) on the contractile response to oxotremorine-M in ileum, trachea, and urinary bladder. In wild-type mice, forskolin caused a significant reduction in the potency of oxotremorine-M in each tissue (i.e., increase in EC₅₀ or decrease in pEC₅₀), and a significant reduction in E_max in the trachea, but not in ileum or urinary bladder. These changes resulted in a significant reduction in the observed coupling efficiency of oxotremorine-M in the three tissues. Similar results were obtained with forskolin in smooth muscle from M₂ receptor knockout mice, although the effects were greater. Forskolin caused 3.6-, 3.1-, and 9.3-fold greater reductions in the coupling efficiency of oxotremorine-M in the ileum, trachea, and urinary bladder of M₂ receptor knockout mice as compared with wild-type mice. These results are summarized in Table 2.

Similar results were obtained with isoproterenol (1.0 µM) except that there was a smaller difference between the relaxant effects of isoproterenol in wild-type and M₂ receptor knockout mice (see Fig. 2). The net effect of isoproterenol was to cause a significant reduction in the coupling efficiency of oxotremorine-M in the three tissues in both wild-type and M₂ receptor knockout mice. Isoproterenol elicited 2.0- and 2.3-fold greater reductions in the observed coupling efficiency of oxotremorine-M in ileum and urinary bladder from M₂ receptor knockout mice as compared with wild-type mice (see Table 3). In contrast, the relaxant effects of isoproterenol in trachea were practically the same in wild-type and M₂ receptor knockout mice (i.e., no significant differences between wild-type and M₂ knockout with regard to the effect of isoproterenol on E_max, shift in EC₅₀, or observed coupling efficiency; P = 0.54, 0.39, and 0.29, respectively). The mean values for the effects of isoproterenol on E_max, EC₅₀, and observed coupling efficiency in the trachea showed greater relaxant effects in wild-type mice compared with M₂ receptor knockout mice, although these differences between groups were not statistically significant. These results are summarized in Table 3.

Figure 3 shows the relaxant effects of forskolin (5.0 µM) and isoproterenol (1.0 µM) on the contractile response elicited by KCl (50 mM). Forskolin caused a significant inhibition of KCl-induced contractions in ileum (55%), trachea (84%), and urinary bladder (39%) from wild-type mice. However, in contrast to that observed with oxotremorine-M as the contractile agent, the relaxant effects of forskolin against KCl-induced contractions were not significantly different between wild-type and M₂ receptor knockout mice in ileum (P = 0.78), trachea (P = 0.073), and urinary bladder (P = 0.84). There was a substantial, although statistically not significant, difference in the trachea; this change was in the direction of forskolin having a larger inhibitory effect in wild-type trachea (84%) compared with that from M₂ receptor knockout trachea (76%). Similar results were obtained when isoproterenol was used as the relaxant agent against KCl-induced contractions. In wild-type mice, isoproterenol caused a highly significant inhibition of KCl-induced contractions in ileum (39%), trachea (82%), and urinary bladder (54%). In ileum and urinary bladder, there were no significant differences in the effects of isoproterenol on KCl-induced contractions between wild-type and M₂ receptor knockout mice (P = 0.95 and 0.99 in ileum and urinary bladder, respectively). In trachea, isoproterenol had similar relaxant effects against KCl-induced contractions in the two types of mice; however, the relaxant effect of isoproterenol was slightly yet significantly greater in wild-type mice as compared with M₂ receptor knockout mice (P = 0.039). Thus, in contrast to the results of the experiments in which oxotremorine-M was used as the contractile agent, the relaxant effects of isoproterenol and forskolin against KCl-induced contractions in M₂ receptor knockout mice were equal to or less than those observed in wild-type mice.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Effects of forskolin (5.0 µM) (a) and isoproterenol (1.0 µM) (b) on contractions elicited to KCl (50 mM) in ileum, trachea, and urinary bladder from wild-type and M2 muscarinic receptor knockout mice. The contractions are expressed as a percentage of the contraction elicited to KCl in the absence of the relaxant agents. The data represent the mean values ± S.E.M. from four wild-type and four knockout mice. *, significantly different from the relaxant effect observed in wild-type mice, P = 0.039.

We also investigated the effects of the potassium channel activator, pinacidil, on the contractile response to oxotremorine-M in smooth muscle from wild-type and M₂ receptor knockout mice (see Table 4). Pinacidil caused a decrease in the potency of oxotremorine-M in ileum, trachea, and urinary bladder and a small increase in E_max in both wild-type and M₂ receptor knockout mice. However, there were no significant differences between the effects of pinacidil in wild-type and M₂ receptor knockout mice.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our observations on the contractile action of the selective muscarinic agonist, oxotremorine-M, in trachea and urinary bladder of M₂ muscarinic receptor knockout mice are similar to those reported by Stengel et al. (2000) using the nonselective cholinergic agonist carbachol. We found that the potency of oxotremorine-M decreased 1.21- and 1.48-fold in trachea and urinary bladder from M₂ receptor knockout mice, respectively, compared with wild-type, whereas Stengel et al. (2000) observed 1.9- and 1.6-fold reductions in potency in these tissues with carbachol. We also observed a significant 2.1-fold reduction in the potency of oxotremorine-M in ileum in M₂ receptor knockout mice compared with wild-type. These results suggest that, in wild-type mice, the M₂ receptor plays some modulatory roles, although the M₃ receptor is capable of eliciting most of the direct contractile response to muscarinic agonists in these tissues. Moreover, Matsui et al. (2002) showed that the direct muscarinic contractile response is completely eliminated in ileum and urinary bladder from mutant mice lacking both M₂ and M₃ receptors. These results show that the postjunctional effects of muscarinic agonists can be attributed almost entirely to M₂ and M₃ muscarinic receptors.

Prior studies on gastrointestinal (Candell et al., 1990; Zhang and Buxton, 1991; Sawyer and Ehlert, 1998), tracheal (Yang et al., 1991), and urinary bladder smooth muscle (Noronha-Blob et al., 1989) have shown that the M₂ muscarinic receptor mediates an inhibition of adenylyl cyclase. In ileum and trachea, this effect has been shown to be prevented by pertussis toxin treatment (Sankary et al., 1988; Thomas and Ehlert, 1994). In contrast, pertussis toxin treatment has no inhibitory effect on M₃ receptor-mediated phosphoinositide hydrolysis in gastrointestinal or tracheal smooth muscle, or on the contractile response to muscarinic agonists in guinea pig ileum, colon, or trachea (Thomas and Ehlert, 1994, 1996; Ostrom and Ehlert, 1999; Sawyer and Ehlert, 1999a,b). This result suggests that the M₂ receptor has little role in mediating the potent contractile response to muscarinic agonists in guinea pig smooth muscle, because uncoupling M₂ receptor signaling with pertussis toxin does not inhibit contraction. However, pertussis toxin treatment has been shown to enhance the relaxant effects of forskolin on oxotremorine-M-mediated contractions in ileum and trachea (Ostrom and Ehlert, 1997, 1998). Presumably, in the presence of forskolin, part of the contractile mechanism of oxotremorine-M involves an M₂ receptor-mediated inhibition of the relaxant effect of forskolin. This mechanism probably involves an M₂ receptor-mediated inhibition of adenylyl cyclase because forskolin is thought to elicit relaxation in smooth muscle through cAMP (see review by Berridge, 1975). By uncoupling this M₂ mechanism with pertussis toxin, the muscarinic contractile response is now more susceptible to inhibition by forskolin. This interpretation is consistent with the observation that pertussis toxin is without effect on forskolin-mediated inhibition of histamine-induced contractions (Ostrom and Ehlert, 1997, 1998). Since histamine elicits contraction in smooth muscle through activation of H₁ receptors (Black et al., 1972), which signal through G_q without activation of G_i (Arrang et al., 1995), one would not expect pertussis toxin to influence histamine-induced contractions or their inhibition by forskolin. Our results in M₂ KO mice with forskolin are consistent with our previous work in guinea pigs using pertussis toxin to inactivate M₂ receptor signaling.

We have also observed that pertussis toxin treatment enhances the relaxant effect of isoproterenol against oxotremorine-M-induced contraction of the guinea pig ileum but not those elicited in the trachea (Thomas and Ehlert, 1994; Ostrom and Ehlert, 1997, 1998). As observed with forskolin, pertussis toxin treatment has no effect on the ability of isoproterenol to inhibit histamine-induced contraction in guinea pig ileum and trachea. Our results on M₂ receptor knockout mice are consistent with these observations. We observed a significant increase in the relaxant action of isoproterenol against oxotremorine-M-induced contraction in ileum and urinary bladder from M₂ receptor knockout mice, but not in trachea. Collectively, our results suggest that pertussis toxin is a useful tool for exploring the role of M₂ receptors in smooth muscle.

It seems unlikely that the increased relaxant effectiveness of forskolin and isoproterenol in M₂ muscarinic receptor knockout mice is due to an increase in the sensitivity of smooth muscle to relaxant agents. Our experiments utilizing KCl as the contractile stimulus showed that there was no increase in the sensitivity of ileum, trachea, or urinary bladder to the relaxant effects of forskolin and isoproterenol in M₂ receptor knockout mice compared with wild-type. Moreover, we found that there was no increase in the relaxant action of the potassium channel activator, pinacidil, against oxotremorine-M-induced contraction in M₂ receptor knockout mice. Since the relaxant mechanism of pinacidil does not involve cAMP, one would not expect to observe a change in the ability of pinacidil to inhibit oxotremorine-M-induced contractions in M₂ receptor knockout mice. Thus, our studies with KCl and pinacidil in M₂ receptor knockout mice are consistent with the postulate that the increased relaxant action of forskolin and isoproterenol against oxotremorine-M-induced contractions is due to the loss of M₂ receptors and not to an increase in the relaxant effectiveness of forskolin and isoproterenol.

The pattern of changes in M₂ muscarinic receptor knockout mice observed in this study shows close agreement with prior pharmacological studies in wild-type guinea pigs. One powerful method for investigating the function of M₂ receptors in wild-type smooth muscle involves inactivating all non-M₂ muscarinic receptors with the aziridinium ion of 4-DAMP mustard (N-(2-chloroethyl)-4-piperidinyl diphenylacetic acid). After this treatment, the muscarinic contractile response is measured in the presence of histamine and forskolin. Under these conditions, the contractile response to oxotremorine exhibits an M₂ profile for pharmacological antagonism in the ileum and trachea (Thomas et al., 1993; Thomas and Ehlert, 1996). Presumably, the contractile mechanism of oxotremorine-M under these conditions involves an M₂ receptor-mediated inhibition of the relaxant effect of forskolin on histamine-induced contractions. When isoproterenol is used in the paradigm in place of forskolin in the ileum, the potency of oxotremorine-M is less, and the pharmacological antagonism exhibits a profile midway between M₂- and M₃-like (Thomas et al., 1993). Analogous types of experiments with isoproterenol in guinea pig urinary bladder (Hegde et al., 1997) and with isoproterenol and forskolin in guinea pig Taenia caeci (Shen and Mitchelson, 1998) have yielded similar results. These data suggest that the M₂ receptor has a smaller role in inhibiting the relaxant effects of isoproterenol as compared with those of forskolin. Moreover, in the trachea, this type of experimental paradigm shows no role for the M₂ receptor in opposing the relaxant effects of isoproterenol on histamine-induced contractions (Ostrom and Ehlert, 1998, 1999). These prior pharmacological studies are consistent with the present data on M₂ receptor knockout mice. As described above, we found a greater increase in the relaxant effectiveness of forskolin compared with isoproterenol against oxotremorine-M-induced contractions in ileum from M₂ receptor knockout mice as compared with wild-type. Also, we found no difference between trachea from wild-type and M₂ receptor knockout mice with regard to the relaxant effectiveness of isoproterenol against oxotremorine-M-induced contractions. These results indicate that the M₂ receptor does not oppose isoproterenol-induced relaxation in the trachea.

It has been suggested that the lack of the role of the M₂ receptor in opposing isoproterenol-induced relaxation in the trachea implies that the relaxant mechanism of isoproterenol in the trachea is through a non-cAMP mechanism (Ostrom and Ehlert, 1998, 1999). Torphy (1994) has previously proposed a non-cAMP mechanism, which could involve stimulation of Ca²⁺-activated K⁺ channels. In bovine trachea, it has been shown that muscarinic receptor activation inhibits the increase in cAMP levels elicited by both forskolin and isoproterenol (Ostrom and Ehlert, 1998). Also, when cAMP levels and relaxation of muscarinic agonist-induced contractions are measured under identical conditions, forskolin-induced relaxation obeys a saturable, increasing function of the cytosolic concentration of cAMP, suggesting a functional relationship between the two (Ostrom and Ehlert, 1998). In contrast, no functional relationship between relaxation and the cytosolic concentration of cAMP was observed for isoproterenol (Ostrom and Ehlert, 1998). An increase in relaxation from 30 to 100% occurred with no change in the cAMP concentration. The smaller role of the M₂ receptor in opposing isoproterenol-induced relaxation in the ileum compared with that caused by forskolin suggests that some, but not all, of the relaxation caused by isoproterenol in the ileum is mediated through cAMP. This could be explained if the G_s activated by -adrenergic receptors acts through at least two pathways, adenylyl cyclase and another pathway that is unopposed by M₂ receptor activation. Forskolin, which acts downstream at the level of adenylyl cyclase, would be unable to elicit such a non-cAMP-dependent mechanism for relaxation.

Prior studies in guinea pigs have shown that, in addition to opposing cAMP-mediated relaxation, M₂ receptor activation also causes a conditional potentiation of M₃ muscarinic receptor-mediated contractions (Sawyer and Ehlert, 1999b). This conclusion was based on the unique pattern of sensitivity of the muscarinic contractile response in the colon to pertussis toxin and both competitive and irreversible muscarinic antagonists. The data were consistent with the postulate that M₂ receptors have no direct contractile action by themselves, but cause a conditional potentiation of M₃ receptor-mediated contractions. As suggested previously (Ehlert, 2003; Ehlert et al., 1997, 1999; Sawyer and Ehlert, 1999b), a possible mechanism for this action could be M₂ receptor-meditated stimulation of the nonselective cation conductance or inhibition of Ca²⁺-activated K⁺ channels. Both of these mechanisms are Ca²⁺-dependent; hence, they fulfill the necessary criterion of being conditional upon Ca²⁺ mobilization by the M₃ receptor. At first, it might seem that a loss of these M₂ responses could account for the increased susceptibility of the muscarinic contractile response to relaxant agents in M₂ receptor knockout mice. However, our method of analysis makes this possibility unlikely. Our null method approach compares equivalent contractile stimuli in the presence and absence of the relaxant agents. There is no inherent reason to assume that a direct contractile stimulus mediated through an interaction between M₂ and M₃ receptors in wild-type mice would be more resistant to relaxant agents than an equivalent stimulus activated only by M₃ receptors in M₂ receptor knockout mice. Moreover, if loss of the latter mechanism were responsible for the increased relaxant action of forskolin and isoproterenol in M₂ receptor knockout mice, then, for a given tissue, we would have expected to observe the same increase in the relaxant effectiveness of both forskolin and isoproterenol. However, this was not observed as described above. In contrast, if the M₂ receptor mediates an attenuation of the relaxant stimulus directly (e.g., through inhibition of adenylyl cyclase), this mechanism could explain why the relaxant activity of forskolin and isoproterenol increased differentially in tissues from M₂ receptor knockout mice. Finally, it is important to mention that our null method also controls for compensatory changes in the activity of M₃ receptors in M₂ receptor knockout mice. Future studies on M₂ receptor knockout mice may help to sort out the mechanism of the M₂ receptor-mediated potentiation of M₃ receptor-mediated contractions.