Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

There are five muscarinic acetylcholine receptor subtypes, designated M₁, M₂, M₃, M₄, and M₅, widely expressed in the central nervous system and in the periphery. Muscarinic receptors modulate the activity of many neurotransmitter systems in the brain, play a key role in memory and learning, and regulate a great number of sensory, motor, and autonomic processes. Moreover, central cholinergic receptor dysfunction has been suggested to be involved in the pathophysiology of schizophrenia (Haroutunian et al., 1994), Alzheimer's disease (Avery et al., 1997; Rodriguez-Puertas et al., 1997), Parkinson's disease (Griffiths et al., 1994; Asahina et al., 1998), and depression (Riemann et al., 1994). Muscarinic agonists have been reported to have antipsychotic-like activity in animal models (Bymaster et al., 1998) and schizophrenic patients (Pfeiffer and Jenney, 1957), to improve cognitive function and psychotic-like behaviors in Alzheimer's disease (Avery et al., 1997; Bodick et al., 1997), and to have antidepressant properties (Janowsky et al., 1972).

Muscarinic M₁ to M₄ receptors are widely distributed throughout the brain, with most brain regions expressing several different muscarinic receptor subtypes (Levey et al., 1991). This overlapping receptor expression pattern, along with a lack of subtype-selective muscarinic agonists and antagonists available for study, has made the functional roles for individual muscarinic receptors difficult to determine. Recently, however, mice lacking specific muscarinic receptor subtypes have been produced by gene ablation technology (Hamilton et al., 1997; Gomeza et al., 1999a,b; Yamada et al., 2001a,b). These genetic knockout (KO) mice have been useful in assigning specific pharmacological functions to individual muscarinic receptors. For example, studies with muscarinic M₂ KO mice have shown that muscarinic M₂ receptors, which are negatively coupled to adenylate cyclase, mediate bradycardia, tremor, hypothermia, and analgesic effects (Gomeza et al., 1999a; Stengel et al., 2000; Bymaster et al., 2001; Gomeza et al., 2001). Mice lacking muscarinic M₄ receptors, which are also negatively coupled to adenylate cyclase, showed increased basal locomotor activity and enhanced dopamine D₁ receptor-mediated locomotor stimulation, suggesting that M₄ receptors exert inhibitory control over dopamine D₁ receptor-dependent locomotor stimulation (Gomeza et al., 1999b, 2001). Muscarinic M₄ receptors have also been shown to lack a functional role in muscarinic agonist-induced tremor and hypothermia and do not play a predominant role in muscarinic agonist-induced analgesia and salivation (Gomeza et al., 1999b; Bymaster et al., 2001).

Muscarinic agonists have been shown to stimulate the hypothalamic-pituitary-adrenocortical axis (HPA) axis via central corticotropin-releasing hormone release (Hedge and De Wied, 1971; Sithichoke and Marotta, 1978; Calogero et al., 1989; Bugajski et al., 1998). The muscarinic receptor subtype mediating corticotropin-releasing hormone secretion and subsequent stimulation of the HPA axis, however, has not been established. In this study, we investigated muscarinic agonist-induced increases in serum corticosterone concentrations using muscarinic M₂ and M₄ receptor single KO and M_2,4 receptor double KO mice and the corresponding WT control strains. Both the M₂ and M₄ receptors are selectively coupled to G proteins of the G_i family that, at a biochemical level, mediate the inhibition of adenylate cyclase (Wess, 1996). We focused on muscarinic M₂ receptors due to their high density in forebrain regions such as hypothalamus (Levy, 1993) and compared effects to another adenylate cyclase-coupled receptor, the muscarinic M₄ receptor, which also is located in rat forebrain regions (Levy, 1993; Yasuda et al., 1993). We used the muscarinic M₂/M₄ receptor-preferring partial agonist [5R-(exo)]-6-[4-butylthio-1,2,5-thiadiazol-3-yl]-1-azabicyclo-[3.2.1]-octane (BuTAC) (Sauerberg et al., 1998), which has been shown to stimulate corticosterone release in rats (S. K. Hemrick-Luecke, unpublished data), has low parasympathomimetic side effects, and reduced potential for stress-induced changes in animal models (Shannon et al., 1999). Scopolamine, a nonselective muscarinic antagonist, was used to block the increases in corticoid levels produced by BuTAC. We also tested agents known to stimulate the HPA axis via activation of central serotonin and dopamine receptors to demonstrate normal function of the HPA axis in muscarinic M₂, M₄, and M_2,4 receptor KO mice.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

The generation of muscarinic M₂ receptor KO mice (genetic background: 129/J1 × CF1; 50%/50%) and muscarinic M₄ receptor knockout mice (129/SvEv × CF1; 50%/50%) has been described previously by Gomeza et al. (1999a,b). M_2,4 receptor double KO mice [129/J1 (25%) × 129SvEv (25%) × CF1 (50%)] were generated by intermating homozygous M₂ and M₄ receptor KO mice (J. Gomeza and J. Wess, unpublished data). In all experiments, age-matched WT mice of the matching genetic background were used as control animals. Mouse genotyping was carried out by polymerase chain reaction analysis of mouse-tail DNA. Immunoprecipitation studies with subtype-selective antibodies confirmed the lack of M₂ and M₄ receptor protein in the homozygous (/) M₂ and M₄ receptor knockout mice, respectively. Moreover, immunoprecipitation studies indicated that the loss of M₂ or M₄ receptors did not lead to changed expression levels of the remaining muscarinic subtypes (Gomeza et al., 1999a,b).

Age-matched male mice (4-6 weeks old) weighing 20 to 25 g were used for all studies. Mouse colonies were amplified by Taconic Farms (Germantown, NY). Mice were housed five per cage in an environmentally controlled room at 22°C with lights on from 7:00 AM to 7:00 PM and were acclimatized for 1 week before experimentation. Food and water were freely available at all times. Animal use protocols were reviewed and approved by institutional animal care and use committees accredited by Assessment and Accreditation Association of Laboratory Animal Care International. BuTAC and pergolide were provided by Lilly Research Laboratories (Indianapolis, IN) and dissolved in 0.01 N HCl for injection. Scopolamine hydrochloride was obtained from Sigma-Aldrich (St. Louis, MO) and dissolved in sterile H₂O for injection. Quipazine [2-(1-piperazinyl) quinoline maleate] was purchased from Miles Laboratories (Elkhart, IN) and dissolved in sterile H₂O for injection. 8-Hydroxy-2-dipropylaminotetralin (8-OH-DPAT) was purchased from Sigma/RBI (Natick, MA) and dissolved in 0.01 N HCl for injection. All mice were administered drugs or vehicles at 5 ml/kg by routes indicated. Mice were sacrificed by decapitation (between 9:00 and 11:00 AM) 1 h after administration of BuTAC, pergolide, 8-OH-DPAT, or quipazine. Trunk blood was collected and allowed to clot, and serum was obtained by centrifugation and stored frozen at 70°C before assay. Serum corticosterone concentrations were measured by radioimmunoassay (corticosterone ³H-RIA kit; ICN Biomedicals, Costa Mesa, CA). Samples were diluted according to kit instructions and analyzed in duplicate. Analysis of variance, followed by Tukey's honestly significant difference test post hoc, was performed with significant differences at P  0.05. Fifty percent effective doses (ED₅₀) were calculated using best-fit linear regression analyses.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Preliminary data showed that BuTAC, a muscarinic M_2,4-preferring partial agonist, administered at a dose of 0.1 mg/kg s.c., significantly increased serum corticosterone concentrations in WT mice as well as in muscarinic M₄ KO mice at 1 h, but did not affect levels of corticoids in muscarinic M₂ KO mice (data not shown). Because doses of BuTAC lower than 0.1 mg/kg s.c. would be expected to be ineffective on corticoid levels in M₂ and M_2,4 KO mice, higher doses were administered, whereas doses of BuTAC administered to WT and M₄ KO mice were chosen to show a dose-dependent effect on corticoid levels.

The effects of BuTAC on serum corticosterone concentrations in muscarinic M₂ receptor KO mice and corresponding WT control mice are shown in Fig. 1A. One hour after BuTAC administration, corticosterone concentrations were dose dependently increased in WT mice at doses of 0.03, 0.1, and 0.3 mg/kg s.c. In muscarinic M₂ KO mice, BuTAC had no effect on corticosterone concentrations at doses up to 1 mg/kg s.c. In contrast, Fig. 1B shows that BuTAC produced nearly identical increases in serum corticosterone concentrations in WT compared with M₄ KO mice, with significant increases occurring at 0.03, 0.1, and 0.3 mg/kg. In M_2,4 double KO mice (Fig. 1C), doses of BuTAC as high as 1 mg/kg had no effect on corticoid levels, whereas BuTAC significantly increased corticosterone concentrations in WT mice.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Effect of BuTAC on mouse serum corticosterone concentrations in muscarinic KO and WT mice. Mouse serum corticosterone concentrations were measured in M2 KO mice (open circles) and corresponding WT mice (closed circles) (A), M4 KO mice (open circles) and corresponding WT mice (closed circles) (B), and M2,4 KO mice (open circles) and corresponding WT mice (closed circles) (C) 1 h after BuTAC administration (0.1 mg/kg s.c.). Means and standard errors for five mice per group are shown (*, P  0.05 compared with vehicle controls).

The effect of the nonselective muscarinic receptor antagonist scopolamine (1 mg/kg) on the increase in corticoid levels produced by BuTAC (0.1 mg/kg) in muscarinic M₄ KO mice and corresponding WT controls is shown in Table 1. The differences in basal levels of corticosterone in vehicle-treated mice are due to normal variability in corticoid levels, because we observed no consistent trend for increased levels in WT compared with muscarinic M₄ KO mice. Scopolamine alone had no significant effect on corticoid levels, but prevented the 5-fold increase in corticoid levels in WT mice and the 10-fold increase in muscarinic M₄ KO mice produced by BuTAC.

The effects of BuTAC and compounds known to increase corticoid levels by other receptor mechanisms were compared in WT and KO mice (Fig. 2). As described above, BuTAC (0.1 mg/kg) significantly increased corticosterone concentrations in M₄ KO mice and in all WT mice, but had no effect on corticoid levels in M₂ KO and M_2,4 KO mice. The dopamine D₁/D₂ receptor agonist pergolide (0.3 mg/kg i.p.) produced significant increases in corticoid levels in M₂ KO mice, M₄ KO mice, and M_2,4 KO mice similar to increases observed in the corresponding WT control mice. Likewise, the serotonin (5-HT)_1A receptor agonist 8-OH-DPAT (0.3 mg/kg s.c.) and the 5-HT_2A receptor agonist quipazine (10 mg/kg s.c.) produced similar increases in corticosterone concentrations in M₂ KO mice, M₄ KO mice, M_2,4 KO mice, and corresponding WT control mice.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Effect of BuTAC, pergolide, 8-OH-DPAT, and quipazine on serum corticosterone concentrations in M2 KO (A), M4 KO (B), M2,4 KO (C), and corresponding WT mice. BuTAC (0.1 mg/kg s.c.), pergolide (0.3 mg/kg i.p.), 8-OH-DPAT (0.3 mg/kg s.c.), and quipazine (10 mg/kg s.c.) were administered to groups of mice (n = 5-10) 1 h before serum was collected. Data are expressed as the percentage of basal levels for vehicle-treated mice (basal levels correspond to 100%). Corticosterone concentrations in control mice were 52.4 ± 10.3 ng/ml for WT and 64.8 ± 20.8 ng/ml for M2 KO, 81.4 ± 25.2 ng/ml for WT and 76.0 ± 21.0 ng/ml for M4 KO, and 72.2 ± 27.8 ng/ml for WT and 60.7 ± 30.6 ng/ml for M2,4 mice. *, P  0.05 compared with control corticoid levels.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The partial muscarinic agonist BuTAC was used to determine muscarinic receptor-mediated effects on corticoid levels in muscarinic KO and WT mice, primarily due to its reduced potential to cause stress-induced changes in corticoid levels due to parasympathetic side effects (Shannon et al., 1999). BuTAC also increases serum corticosterone levels in rats (S. K. Hemrick-Luecke, unpublished observation) and mice at doses shown to be pharmacologically active (Sauerberg et al., 1998; Rasmussen et al., 2001). BuTAC is a partial agonist at muscarinic M₂ receptors (K_i = 0.4 nM) and M₄ receptors (K_i = 0.6 nM), and an antagonist at muscarinic M₁ receptors (K_i = 0.6 nM), M₃ receptors (K_i = 0.2 nM) and M₅ receptors (K_i = 0.7 nM) in cloned cell lines (Sauerberg et al., 1998). In radioligand binding assays BuTAC at 1 µM did not appreciably interact with other nonmuscarinic neuronal receptors or neurotransmitter transporters with the exception of slight affinity for -receptors in vitro (Sauerberg et al., 1998). At doses similar to those increasing corticoid levels in WT mice, BuTAC inhibited conditioned avoidance responding in rats without response failure or catalepsy and displayed large separations between antidopaminergic doses and doses that produce parasympathomimetic effects in mice (Sauerberg et al., 1998; Shannon et al., 1999).

BuTAC produced dose-dependent increases in serum corticosterone concentrations with similar potency and magnitude in all strains of WT control mice and muscarinic M₄ receptor KO mice. This BuTAC-induced increase in corticoid levels in M₄ KO and WT mice was blocked by scopolamine, a nonselective antagonist with high affinity for muscarinic receptor subtypes (Bolden et al., 1992), indicating that stimulation of the HPA axis by BuTAC was due to muscarinic receptor activation. In muscarinic M₂ receptor KO mice and in M_2,4 receptor KO mice, however, BuTAC had no significant effect on corticoid levels at doses 10 to 30 times higher than doses required to significantly increase corticosterone concentrations in WT control mice. The muscarinic M₂ receptor subtype thus regulates corticoid release in mice. The location of these receptors, however, cannot be determined from these data. Further studies are needed to elucidate whether muscarinic agonist-induced corticoid release is mediated by centrally located muscarinic M₂ receptors that stimulate corticotropin-releasing hormone-containing neurons in the hypothalamus, resulting in corticotropin-releasing factor, adrenocorticotropin hormone, and corticosterone release or by peripheral stimulation of muscarinic M₂ receptors located on the adrenal cortex. Nonetheless, stimulation of the HPA axis is mediated by muscarinic M₂ receptors in mice.

We also show that the dopamine D₂ receptor agonist pergolide, the 5-HT_1A receptor agonist 8-OH-DPAT, and the 5-HT_2A receptor agonist quipazine all produced increases in corticosterone concentrations in M₂, M₄, and M_2,4 KO mice, similar to the increases observed in corresponding WT control mice. This suggests that the HPA axis is able to function normally in muscarinic M₂, M₄, and M_2,4 receptor KO mice with regard to dopaminergic and serotonergic receptor agonist-induced stimulation. Of particular interest is that the 5-HT_1A receptor agonist 8-OH-DPAT produced an increase in corticosterone concentrations in the muscarinic M₂ and M_2,4 receptor KO mice similar to WT mice. 5-HT_1A receptors, like muscarinic M₂ and M₄ receptors, are negatively coupled to adenylate cyclase. Similar increases in 8-OH-DPAT-induced corticoid levels in WT mice and muscarinic KO mice indicate an intact adenylate cyclase transduction system. Thus, the only abnormality observed on corticoid levels is the lack of effect by the muscarinic M_2,4 receptor-preferring agonist BuTAC on corticoid levels in mice lacking functional muscarinic M₂ receptors.

Because ligands able to clearly distinguish between muscarinic receptor subtypes are not available, stimulation of the HPA axis may provide a means to evaluate muscarinic M₂ receptor activation in animals and humans. Stimulation of the HPA axis by muscarinic agents would provide a measure of muscarinic M₂ receptor agonist activity, thus allowing for development of drugs with fewer side effects. Muscarinic M₂ receptors have been implicated in the regulation of many central and peripheral functions, including control of heart rate and contraction of smooth muscle (Gomeza et al., 1999a, 2001; Stengel et al., 2000; Bymaster et al., 2001). In the central nervous system, muscarinic M₂ receptors can function as inhibitory autoreceptors (Zhang et al., 1999) implicated in parasympathetic effects, movement control, and changes in body temperature and pain sensitivity (Gomeza et al., 1999a; 2001; Bymaster et al., 2001).

Evidence exists that cholinergic mechanisms may have a regulatory effect in affective disorders. The cholinergic-adrenergic hypothesis of mania and depression (Janowsky et al., 1972) states that hypercholinergic (or hypoadrenergic) activity results in depression, whereas hypocholinergic (or hyperadrenergic) activity results in mania. Cholinergic muscarinic mechanisms have been implicated in preclinical (Hassey and Hannin, 1991; Overstreet et al., 1986, 1998) and clinical (Dilsaver, 1986; Janowsky and Overstreet, 1995) models of depression. Furthermore, cholinomimetics and direct-acting cholinergic agonists increase levels of corticotropin-releasing hormone, adrenocorticotropin hormone, and cortisol (Risch et al., 1983; Janowsky et al., 1986; Suda et al., 1987; Calogero et al., 1989), although evidence linking stimulation of the HPA axis to cholinergic activity in humans is lacking. Nonetheless, depression is associated with hyperactivity of the HPA axis (Nemeroff, 1996; Arborelius et al., 1999), and development of muscarinic acetylcholine M₂ receptor antagonists might be effective in the treatment of depression.

In conclusion, our data show that the muscarinic M₂ receptor subtype mediates muscarinic agonist-induced stimulation of corticosterone release, whereas muscarinic M₄ receptors do not seem to be involved in this activity.