Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Muscarinic receptors are particularly abundant in the cerebral cortex, where they are believed to participate in the cholinergic regulation of arousal, cognitive functions and synaptic plasticity (Bartus et al., 1982; Ashe and Weinberger, 1991). Radioligand binding, molecular genetic and immunological studies have demonstrated that the rat cerebral cortex expresses four distinct muscarinic receptor subtypes (M₁ to M₄), which display different densities and cellular distribution (Buckley et al., 1988; Waelbroeck et al., 1990; Levey et al., 1991). Moreover, rat cortical muscarinic receptor subtypes have been reported to be coupled to distinct signal transduction mechanisms. Thus, M₁ and M₃ receptors have been found to stimulate phosphoinositide hydrolysis, whereas receptors pharmacologically equivalent to the m4 gene product have been found to inhibit cyclic AMP production (Forray and El-Fakahany, 1990; McKinney et al., 1991).

Because one of the cellular functions of muscarinic receptors is the modulation of the responsiveness of cortical cells to incoming inputs (Ashe and Weinberger, 1991; Cox et al., 1994), it is important to understand how muscarinic signaling integrates with and regulates other neurotransmitter stimuli. The identification of receptor interactions may be exploited for the development of therapeutic strategies aimed at modulating synaptic transmission at specific sites.

In the present study, we report that in the rat frontal cortex, the activation of muscarinic receptors potentiates the stimulation of cyclic AMP formation elicited by CRH, a neurotransmitter/neuromodulator involved in the regulation of stress responses and of learning and memory processes (Koob and Bloom, 1985). The pharmacological profile of the muscarinic receptors mediating the potentiation of the CRH response resembles that of the M₁ receptor subtype and is distinct from that displayed by muscarinic receptors coupled to inhibition of adenylyl cyclase activity. Moreover, unlike the inhibitory response, the signaling mechanism leading to adenylyl cyclase potentiation involves pertussis toxin-insensitive G proteins.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. [-³²P]ATP (30-40 Ci/mmol), [2,8-³H]cyclic AMP (25 Ci/mmol) and [³⁵S]GTPS (1306 Ci/mmol) were purchased from Du Pont de Nemours (Bad Homburg, Germany). Human/rat CRH was obtained from Peninsula Laboratories Inc. (Merseyside, England). Forskolin was from Calbiochem (La Jolla, CA). Unlabeled GTPS was from Boehringer (Mannheim, Germany). Pirenzepine and AF-DX 116 were obtained from Dr. Karl Thomae GmbH (Biberach an der Riss, Germany), whereas AQ-RA 741 was from Boehringer Ingelheim (Milan, Italy). Oxotremorine methiodide (oxotremorine-M), telenzepine, HHSiD, pFHHSiD, methoctramine and pertussis toxin were purchased from Research Biochemical Inc. (Natick, MA). MT-1 toxin isolated from Dendroaspis angusticeps was obtained from Alomone Labs (Jerusalem, Israel). Himbacine and R-trihexyphenidyl were donated by Prof. W. C. Taylor, University of Sidney (Sidney, Australia), and Prof. A. J. Aasen, University of Oslo (Oslo, Norway), respectively. Staurosporine was generously provided by Kyowa Hakko Kogyo Co. (Tokyo, Japan). U-73122 (1-(6-((17-3-methoxyestra-1,3,5(10)-trien-17-yl)amino)hexyl)1H-pyrrole-2,5-dione) was from Biomol (Hamburg, Germany). ACh, methacholine, oxotremorine, carbachol, indomethacin, nordihydroguaiaretic acid and the other chemicals were from Sigma Chemical Co. (St. Louis, MO).

Membrane preparation. Male Sprague-Dawley rats (250-350 g) were sacrificed by decapitation and the brains rapidly removed from the skulls. The brains were transferred to an ice-cold plate and placed on their dorsal surfaces. After removal of the olfactory bulbs and tracts, a cut was made in the coronal plane at the level just anterior to the olfactory tubercles. The frontal lobes were homogenized by using a Teflon-glass tissue grinder (12 up-and-down strokes by hand) in 10 volumes (w/v) of ice-cold buffer containing 10 mM HEPES-NaOH, 1 mM EGTA, 1 mM DTT and 1 mM MgCl₂ (pH 7.4). The homogenate was diluted with the same medium and centrifuged at 27,000 × g for 20 min at 4°C. The pellet was resuspended in the same buffer and centrifuged as above. The final pellet was resuspended to a final protein concentration of 1.5 mg/ml and either used immediately for the adenylyl cyclase assay or stored at 70°C.

Adenylyl cyclase assay. The enzyme activity was assayed in a 100-µl reaction mixture containing 50 mM HEPES-NaOH buffer (pH 7.4), 2.3 mM MgCl₂, 0.1 mM [-³²P]ATP (70 cpm/pmol), 1 mM [³H]cyclic AMP (80 cpm/nmol), 0.3 mM EGTA, 1.3 mM DTT, 1 mM 3-isobutyl-1-methylxanthine, 5 mM phosphocreatine, 50 U/ml creatine kinase, 10 µM GTP, 50 µg of BSA, 10 µg of bacitracin and 10 kallikrein inhibitor units of aprotinin. Eserine (10 µM) was included when the effect of ACh was examined. The incubation was started by the addition of the tissue preparation and was carried out at 30°C for 10 min. [³²P]cyclic AMP was isolated according to the method of Salomon et al. (1974). When the forskolin-stimulated enzyme activity was determined, the reaction mixture contained 10 µM forskolin, 0.05 mM [-³²P]ATP and 100 µM GTP. Assays were performed in duplicate.

[³⁵S]GTPS binding assay. The binding of [³⁵S]GTPS was assayed in a reaction mixture (final volume, 100 µl) containing 25 mM HEPES-NaOH (pH 7.4), 150 mM NaCl, 4 mM MgCl₂, 1 mM EGTA, 1 mM DTT, 50 µM GDP, 10 µM eserine and 2.5 nM [³⁵S]GTPS. The incubation was started by the addition of the membrane suspension (4-6 µg of protein) and was performed at 30°C for 60 min. Incubation was terminated by adding 5 ml of ice-cold buffer containing 10 mM HEPES-NaOH (pH 7.4) and 1 mM MgCl₂, immediately followed by rapid filtration through glass-fiber filters (Whatman GF/C) presoaked in the same buffer. The filters were washed two times with 5 ml of buffer, and the radioactivity trapped was determined by liquid scintillation spectrometry. Nonspecific binding was determined in the presence of 100 µM GTPS. Assays were performed in duplicate.

Intracerebral injection of pertussis toxin. Pertussis toxin was dissolved in a solution containing 50 mM sodium phosphate buffer and 250 mM NaCl (pH 7.4). The animals were anesthetized with chloral hydrate (400 mg/kg i.p.) and placed in a stereotaxic frame. The toxin (3.0 µg in 6 µl) was injected bilaterally into the frontal cortex at a rate of 0.3 µl/min according to the following coordinates: A + 4.0, L ± 2.0 and V  3.0 with the bregma as zero (Paxinos and Watson, 1982). Control animals received an equal volume of vehicle containing 3 µg of BSA. The animals were sacrificed 5 days after surgery, and membranes were prepared from vehicle- and toxin-treated frontal cortices. Three tissue preparations were investigated. Protein content was determined by the method of Bradford (1976), using BSA as a standard.

Statistical analysis. Results are expressed as means ± S.E. values. Data from concentration-response curves were analyzed by a least-squares curve-fitting computer program (Graph Pad Prism, ISI Software, San Diego, CA). Agonist concentrations producing half-maximal effects (EC₅₀ values) were converted to the logarithmic form (pEC₅₀ = negative logarithm of EC₅₀), because these values are log-normally distributed (Fleming et al., 1972). Antagonist pA₂ values were calculated from Arunlakshana-Schild regressions (Arunlakshana and Schild, 1959), in which the log of dose ratios  1 is plotted as a function of the antagonist concentration. In experiments examining the effects of a single concentration of antagonist, the inhibition constant (K_i) was calculated according to the equation: EC₅₀b = EC₅₀ a (1+ I/K_i), where EC₅₀ a and EC₅₀ b are the concentrations of the agonist producing a half-maximal effect in the absence and presence of the antagonist, respectively, and I is the concentration of the antagonist. The K_i values were converted to the logarithmic form (pK_i). Statistically significant differences between concentration-response curves were determined by two-way analysis of variance with repeated measures. The statistical significance of the difference between means was determined by Student's t test.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of cholinergic agonists on CRH-stimulated adenylyl cyclase activity. As shown in figure 1, in membranes of the rat frontal cortex, CRH caused a concentration-dependent stimulation of adenylyl cyclase activity, with a maximal effect corresponding to a 71.5 ± 5.6% increase of basal adenylyl cyclase activity (P < .001, n = 6). The addition of either ACh (10 µM and 1 mM) or oxotremorine-M (10 µM), a muscarinic agonist, potentiated the stimulatory effect of the peptide. At 0.5 to 1.0 µM CRH, the increase of adenylyl cyclase activity elicited by the peptide was enhanced by 26.7 ± 1.9% (P < .05) and 53.1 ± 3.5% (P < .001) in the presence of 10 µM and 1 mM ACh, respectively, and by 49.8 ± 2.5% (P < .001) in the presence of 10 µM oxotremorine-M. ACh and oxotremorine-M did not significantly change the pEC₅₀ value of CRH (control, 7.42 ± 0.05; ACh 1 mM, 7.55 ± 0.06; P > .05, n = 4; oxotremorine-M, 7.59 ± 0.09; P > .05, n = 3).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Potentiation of CRH stimulation of adenylyl cyclase activity by ACh in membranes of rat frontal cortex. The enzyme activity was assayed at the indicated concentrations of CRH in the absence () and presence of 10 µM () and 1 mM () ACh and 10 µM oxotremorine-M (). Data represent the means ± S.E. of four (1 mM ACh) and three (10 µM ACh and oxotremorine-M) experiments, each performed on a separate membrane preparation. P < .01 for the difference between control and either 1 mM ACh or 10 µM oxotremorine-M curves and P < .05 for the difference between control and 10 µM ACh curves by analysis of variance.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effects of cholinergic agonists on adenylyl cyclase activity of rat frontal cortex. A, The enzyme activity was assayed in the absence (open symbols) and presence (closed symbols) of 0.5 µM CRH at the indicated concentrations of ACh (,), methacholine (,) and oxotremorine (,). B, The enzyme activity was assayed as in A at the indicated concentrations of carbachol (,), oxotremorine-M (,) and arecoline (,). Data are the means ± S.E. of four experiments for each agonist.

Antagonism of ACh potentiation. As shown in figure 3, the concentration-response curve of ACh in potentiating the CRH-stimulated adenylyl cyclase activity was progressively shifted to the right by increasing concentrations of the M₁ antagonist pirenzepine. A Schild plot of pirenzepine antagonism yielded a pA₂ value of 7.90 and a slope value of 0.93. Telenzepine and R-trihexyphenidyl, two other receptor antagonists known to bind with high affinity to M₁ receptors (Doods et al., 1987; Dorje et al., 1991), were also quite potent, with pA₂ values of 8.10 and 8.19, respectively (table 1). These values were not significantly different (P > .05) from that of pirenzepine. Conversely, the M₂-preferring antagonists AQ-RA 741 and AF-DX 116 (Dorje et al., 1991; Caufield, 1993) displayed significantly (P < .001) lower potencies, with pA₂ values of 6.44 and 7.00, respectively. Similarly, methoctramine, an additional M₂-preferring antagonist, and himbacine, which binds to M₂ and M₄ receptors with higher affinity than to M₁ and M₃ receptors (Caufield, 1993), showed pK_i values of 7.56 and 7.41, respectively (table 1). The M₃-preferring compounds HHSiD and pFHHSiD (Caufield, 1993) antagonized the ACh facilitatory effect, with pA₂ values of 7.98 and 7.83, respectively. MT-1, a peptide toxin that has been reported to bind preferentially to cloned m1 and m4 receptors and to have a much lower affinity for the other subtypes (Adem and Karlsson, 1997), antagonized the ACh effect with a pK_i value of 6.82. Per se, none of the tested muscarinic antagonists affected CRH-stimulated adenylyl cyclase activity.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Antagonism of ACh-induced potentiation of CRH-stimulated adenylyl cyclase activity by pirenzepine. The ACh potentiation of the enzyme activity stimulated by 0.5 µM CRH was determined in the absence () and presence of 15 (), 50 (), 100 () and 500 () nM pirenzepine. The data are expressed as percent of the maximal potentiation observed in the absence of the antagonist and represent the means ± S.E. of four experiments. CRH-stimulated enzyme activity (expressed as pmol cyclic AMP·min·mg protein ± S.E.) were: control, 31.4 ± 1.2; 1 mM ACh, 48.1 ± 2.4. Inset, Schild plot of pirenzepine antagonism.

Muscarinic inhibition of forskolin-stimulated adenylyl cyclase activity. The addition of 10 µM forskolin stimulated adenylyl cyclase activity of the rat frontal cortex by ~5.5-fold. The stimulatory effect of forskolin was inhibited by ACh in a concentration-dependent manner, with a pEC₅₀ value of 5.92 ± 0.06 (fig. 4). The maximal inhibitory effect corresponded to a 21.5 ± 2.5% reduction of control activity (P < .01, n = 12). Carbachol and methacholine inhibited the forskolin-stimulated cyclic AMP formation as effectively as did ACh, with pEC₅₀ values of 5.69 ± 0.07 and 5.41 ± 0.08, respectively. The maximal inhibitory effect elicited by oxotremorine and arecoline corresponded to 78% and 80%, respectively, of that elicited by ACh, with pEC₅₀ values of 6.61 ± 0.06 and 5.53 ± 0.10, respectively (fig. 4).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Inhibition of forskolin-stimulated adenylyl cyclase activity by cholinergic receptor agonists in membranes of rat frontal cortex. The enzyme activity was assayed in the presence of 10 µM forskolin at the indicated concentrations of oxotremorine (), ACh (), carbachol (), methacholine () and arecoline (). Data are the means ± S.E. of at least three experiments for each agonist.

Signal transduction mechanisms. The ACh potentiation of CRH-stimulated adenylyl cyclase activity was not affected by the addition of the phospholipase C inhibitor U-73122 (5 µM) (Bleasdale et al., 1990) and the protein kinase inhibitor staurosporine (100 nM). Also, indomethacin (10 µM) and nordihydroguaiaretic acid (10 µM), two inhibitors of arachidonic acid metabolism via cyclooxygenase and lipoxygenase pathways, respectively, were without effect (results not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Effect of intracerebral injection of pertussis toxin on ACh potentiation of CRH receptor activity (A). The adenylyl cyclase activity stimulated by 1 µM CRH (reported as pmol of cyclic AMP·min·mg protein ± S.E.) was determined in vehicle-() and in pertussis toxin-() treated membranes at the indicated concentrations of ACh. Data are the means ± S.E. of three experiments, each performed on a separate membrane preparation. Effect of intracortical injection of pertussis toxin on Ach inhibition of forskolin-stimulated activity (B). The enzyme activity was assayed in vehicle-() and pertussis toxin-() treated membranes at the indicated concentrations of ACh. The concentration of forskolin was 10 µM. Data are the means ± S.E. of three experiments, each performed on a separate membrane preparation. P < .01 for the difference between the two curves of enzyme response to ACh by analysis of variance.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Effect of intracerebral injection of pertussis toxin on ACh stimulation of [35S]GTPS binding to rat cortical membranes. The binding of [35S]GTPS (2.5 nM) was assayed in vehicle-() and pertussis toxin-() treated membranes at the indicated concentrations of ACh. The inset shows the net increases elicited by the agonist above basal values. Data are the means ± S.E. of three experiments, each performed on a separate membrane preparation. P < .01 for the difference between the responses obtained in control and toxin treated-membranes by analysis of variance.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study shows that in the rat frontal cortex, activation of muscarinic receptors potentiates CRH-stimulated cyclic AMP formation. This interaction results in an amplification of the maximal stimulatory response elicited by CRH, without a significant change in the potency of the peptide. This suggests that muscarinic receptors enhance the signal transduction of CRH receptors rather than increase the affinity of the receptors for the peptide. A similar modulatory effect has previously been described in the rat olfactory bulb, where muscarinic receptors facilitate CRH receptor activity without affecting the binding of ¹²⁵I-CRH (Olianas and Onali, 1993). In the absence of CRH, basal adenylyl cyclase activity is consistently increased only by ACh and carbachol, whereas the other agonists are either completely inactive or exert minor effects. The stimulation of basal adenylyl cyclase activity by ACh and carbachol is, however, modest (12-14% increase) and requires higher agonist concentrations when compared with the potentiation of CRH-stimulated enzyme activity. This indicates not only that muscarinic receptors per se can activate adenylyl cyclase but also that concurrent CRH receptor activation greatly enhances the coupling efficiency of this response.

The analysis of concentration-response curves shows that various cholinergic agonists possess different abilities of enhancing CRH-stimulated adenylyl cyclase activity. Oxotremorine-M, methacholine and carbachol are as effective as Ach, but their potencies vary according to the following rank order: oxotremorine-M > ACh > methacholine > carbachol. Of particular interest is the finding that when compared with ACh, oxotremorine is a very weak agonist and arecoline produces only a modest stimulatory effect. Both the agonist rank order of potency and relative efficacies are quite similar to those described for the muscarinic stimulation of phosphoinositide hydrolysis in the cerebral cortex, a response predominantly mediated by M₁ receptors with a minor contribution by M₃ receptors (Brown et al., 1984; Fisher and Bartus, 1985; Forray and El-Fakahany, 1990; McKinney et al., 1991). In mammalian cortical neurons, oxotremorine has also been reported to be inactive in enhancing cell excitability through M-current inhibition (McCormick and Prince, 1985), another response involving M₁ and M₃ receptors (Caufield, 1993). In cell lines transfected with the genes of the various muscarinic receptor subtypes, oxotremorine and arecoline were found to be full agonists at the m2 and m4 receptors but significantly less effective than carbachol in eliciting m1- and m5-mediated functional responses (Wang and El-Fakahany, 1993).

The possibility that M₁ and M₃ receptors are involved in the muscarinic potentiation of CRH-stimulated adenylyl cyclase activity was investigated by examining the effects of a number of subtype-selective receptor antagonists. The results obtained indicate that the M₁-selective antagonists pirenzepine, telenzepine and R-trihexyphenidyl are more potent blockers than are the M₂ antagonists AF-DX 116, methoctramine and AQ-RA 741 and the M₂ and M₄ antagonist himbacine. In terms of absolute values, the inhibitory constants of these drugs agree with their affinities for the M₁ receptors reported in radioligand binding and functional studies (Caufield, 1993). Although these data indicate the involvement of M₁ rather than M₂ and M₄ receptors, the high potencies displayed by pFHHSiD and HHSiD (pA₂ values of 7.83 and 7.98, respectively) suggest the possible participation of M₃ receptors also. To investigate this point we tested the effect of MT-1, a snake venom peptide that has been found to bind with high affinity to cloned m1 and m4 receptors and with low affinity (K_i > 1000 nM) to the other receptor subtypes (Adem and Karlsson, 1997). MT-1 antagonizes the ACh potentiation of CRH receptor activity with a pK_i of 6.82, which is close to its affinity for the m1 receptor subtype (49 nM, Adem and Karlsson, 1997). Although these data do not rule out the participation of M₃ receptors, they suggest that the M₃ contribution, if present, is quite modest. The high pA₂ values of pFHHSiD and HHSiD could be explained by the fact that these antagonists possess high affinity for M₁ receptors also (Dorje et al., 1991).

Previous studies have reported that in cortical minces, forskolin-stimulated cyclic AMP accumulation is inhibited by the activation of muscarinic receptors (McKinney et al., 1991). It was therefore of interest to see whether this inhibitory response could be detected in a cortical membrane preparation, in which potentiation of CRH receptor activity was observed. Indeed, when cortical adenylyl cyclase is stimulated by forskolin, the addition of cholinergic agonists induces an inhibitory response. However, the pharmacological profile of the inhibitory effect is markedly different from that displayed by the muscarinic potentiation of CRH-stimulated adenylyl cyclase activity. For instance, oxotremorine and arecoline behave almost as full agonists, eliciting a maximal inhibitory response equal to 78% to 80% of that caused by ACh. Moreover, the ACh inhibitory effect is antagonized by methoctramine and himbacine more potently than by HHSiD and pFHHSiD and by pirenzepine with a very low potency. The pK_i values of the antagonists as well as their rank order of potencies are quite close to those reported for either the M₂ or M₄ receptors. Collectively, the data are consistent with the possibility that the cortical muscarinic receptors coupled to the inhibition of forskolin-stimulated adenylyl cyclase activity belong to the M₂ subtype or, as has previously been postulated (McKinney et al., 1991), are m4 gene products. The possible involvement of M₄ receptors may explain the finding that methoctramine blocks the muscarinic inhibitory and stimulatory effects with similar potencies (pK_i values of 7.57 and 7.56, respectively). Indeed, this drug, although effective in discriminating between M₂ and M₃ receptors, poorly distinguishes M₄ from M₁ receptors (Dorje et al., 1991).

A series of experiments were performed to gain information about the signal transduction mechanisms mediating M₁ potentiation of CRH-stimulated adenylyl cyclase activity. Previous studies have shown that the activation of M₁ and M₃ receptors can increase intracellular cyclic AMP levels through multiple mechanisms, including stimulation of phospholipase C and phospholipase A₂, prostaglandin formation, Ca⁺⁺ mobilization, stimulation of Ca⁺⁺/calmodulin-dependent adenylyl cyclase and stimulation of protein kinase C (Felder et al., 1989; Abdel-Latif et al., 1992; Baumgold et al., 1992; Esqueda et al., 1996). Moreover, in pituitary cells and in fetal rat forebrain cell cultures, activation of protein kinase C has been found to enhance the CRH stimulation of cyclic AMP accumulation (Cronin et al., 1986; Kapcala and Aguilera, 1995). The present study, however, shows that both the phospholipase C inhibitor U-73122 and the potent protein kinase inhibitor staurosporine failed to prevent the muscarinic potentiation of CRH signaling in membranes of the rat frontal cortex. The muscarinic effect is also insensitive to inhibitors of arachidonic acid metabolism, such as indomethacin and nordihydroguaiaretic acid. These data suggest that phospholipid breakdown and protein kinase C activation do not mediate the facilitating effect of muscarinic receptors on cyclic AMP formation. Moreover, this response seems to involve the participation of pertussis toxin-insensitive G proteins. In fact, we found that in the rat frontal cortex, toxin treatment impaired the muscarinic inhibition of forskolin-stimulated adenylyl cyclase activity and the ACh-induced stimulation of [³⁵S]GTPS binding to membrane G proteins, a likely result of the uncoupling of muscarinic receptors from G_i/G_o (Spiegel et al., 1992). However, in the same membrane preparations, the muscarinic potentiation of CRH-stimulated adenylyl cyclase activity was largely unaffected when compared with the response obtained in control membranes, indicating that G_i/G_o activation is not required for the response. M₁ receptors are known to couple preferentially to pertussis toxin-insensitive G proteins of the G_q/G₁₁ type (Bernstein et al., 1992), and the subunits released from these G proteins in combination with G_s activated by CRH receptors may stimulate specific forms of adenylyl cyclase (Tang and Gilman, 1992). This possibility is supported by the observation of the expression in the rat cerebral cortex of the -stimulated type II adenylyl cyclase (Furuyama et al., 1993; Mons et al., 1993). However, recent studies have shown that the cloned m1 receptor may interact directly with G_s, which is pertussis toxin-insensitive and stimulates all types of adenylyl cyclase isoforms so far cloned (Burford and Nahorski, 1996). Thus, additional studies are required to identify the nature of the pertussis toxin-insensitive G protein(s) and the molecular mechanism(s) involved in the muscarinic potentiation of CRH receptor activity.

The demonstration of functional interaction between the M₁ and CRH receptors in a cell-free system provides important evidence for the colocalization of the receptors on cellular structures of the frontal cortex, where they control a common pool of adenylyl cyclase. This observation is in line with previous studies showing an interplay between Ach and CRH in the frontal cortex (Crawley et al., 1985; De Souza and Battaglia, 1986). The administration of M₁ receptor agonists is currently considered to be useful for the treatment of cognitive dysfunctions (Elhert et al., 1994). In addition, the potentiation of central CRH transmission has been proposed to be beneficial in the treatment of Alzheimer's disease (Behan et al., 1995). The finding of a positive interaction between M₁ and CRH receptors in the cerebral cortex suggests that the combination of M₁-selective agonists and compounds that increase central CRH receptor activity may elicit more than additive cognitive-enhancing effects.