Substantial experimental evidence now exists to show that G protein-coupled receptors (GPCRs) can productively couple to G proteins in the absence of agonist to produce a measurable downstream response (see de Ligt et al., 2000; Parnot et al., 2002; Milligan, 2003). An early demonstration of constitutive activity was the observation of agonist-independent G protein GTPase activity mediated by δ-opioid receptors endogenously expressed in NG108-15 neuroblastoma glioma cells (Costa and Herz, 1989). However, the majority of GPCRs exhibit low levels of agonist-independent activity even when overexpressed. Therefore, much of the pioneering work on GPCR constitutive activity has exploited either single point mutations (Kjelsberg et al., 1992) or small amino acid sequence replacements (Samama et al., 1993) to create constitutively active receptors that have been presumed to (at least partially) reflect the endogenous, agonist-stabilized activated state. These constitutively active mutants also show both increased affinity and intrinsic activity for agonists. To accommodate these empirical observations, the “extended ternary complex” model (Samama et al., 1993) and the more thermodynamically complete “cubic ternary complex model” (Weiss et al., 1996) were developed. Central to these models is the concept that receptors exist in an equilibrium between inactive (R) and active (R*) receptors. The distribution of the receptor species is governed by an equilibrium constant, which for most endogenously expressed receptors is such that the equilibrium favors R, and consequently, little constitutive activity is seen. The enhanced ability to detect constitutive activity led to the discovery of a unique subclass of antagonists, which exert their actions by actively reducing basal effector activity (Samama et al., 1994). This novel pharmacological property, termed “inverse-agonism,” has been modeled by assuming that inverse agonists preferentially bind to the inactive R or resting state of the receptor (Leff, 1995).

Recent studies where mutated human muscarinic acetylcholine (mACh) receptor subtypes have been expressed in recombinant cell lines have identified specific regions (Spalding et al., 1998) and single amino acid residues (Spalding et al., 1997) that seem to be important in maintaining the inactive conformation of the receptor and whose substitution(s) increase constitutive activity. These observations align with those from other receptors, which suggest that the activation process involves a specific rotation of receptor regions, particularly affecting the relative positions of the TM3 and TM6 domains (Farrens et al., 1996; Javitch et al., 1997). Although receptor-G protein contact sites are as yet poorly defined, it has been suggested that α-helical cytoplasmic extensions of TM5 and TM6 may provide the surface for G protein interaction (Kristiansen, 2004) and that one function of the inactive receptor must be to prevent these activating domains from interacting with the G protein (Kristiansen, 2004). In support of this model, mutagenesis of the C-terminal end of the third intracellular loop in the human M₁ (Hoegger et al., 1995), rat M₃ (Schmidt et al., 2003), and human M₂ (Liu et al., 1996) mACh receptors have been shown to elicit constitutive activity.

In the present study, we have introduced a single point mutation (N514Y) into the human M₃ mACh receptor, homologous to that which generates constitutive activity in the M₅ mACh receptor (Spalding et al., 1997). The locus of this mutation is predicted to lie at the junction of TM6 and the third extracellular loop, and this substitution is predicted to mimic some of the movements involved in the receptor activation process leading to the cytoplasmic exposure of G protein binding domains. We show that the “inverse agonist” behavior of certain mACh antagonists is dependent on the functional assay used. Therefore, the assumption that inverse agonists operate by stabilizing a common inactive conformation of the receptor may be too simplistic.

Materials and Methods

Materials. 4-Diphenylacetoxy-N-methylpiperidine (4-DAMP), atropine, carbamylcholine (carbachol), GTP, methacholine (MCh), methoctramine, oxotremorine, pilocarpine, and poly-l-lysine were from Sigma-Aldrich (Poole, UK). Myo-[2-³H]inositol, N-methyl-[³H]scopolamine ([³H]NMS), [³²P]orthophosphate, and protein A-Sepharose were from Amersham Biosciences Ltd. (Chalfont St. Giles, UK). GeneJuice transfection reagent was from EMD Biosciences (Nottingham, UK). Human embryonic kidney (HEK)-293 cells and the M₃ antibody were a gift from Dr. A. B. Tobin (University of Leicester). All the cell culture reagents were from Invitrogen (Paisley, UK). All the other reagents were of analytical grade.

Mutagenesis and Expression. The M₃ mutant receptor was created using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) using the following primer and its complement: 5′-CCA TAC AAC ATC ATG GTT CTG GTA TAC ACC TTT TGT GAC AGC TGC-3′. Residues in bold encode both an amino acid residue for the N514Y substitution and a unique restriction site for AccI, which allowed for initial confirmation of successful mutagenesis before subsequent sequencing confirmation.

Transient Transfection of HEK-293 Cells. HEK-293 cells were grown in minimum Eagle's medium-α supplemented with 10% fetal calf serum, 100 IU/ml penicillin, 100 μg/ml streptomycin, and 2.5 μg/ml amphotericin B. Cells were maintained at 37°C in 5% CO₂/humidified air. Transfections were performed on cells grown to 40 to 50% confluence using the transfection reagent GeneJuice (optimized at 1 μg of DNA/4 μl of GeneJuice) according to the manufacturer's protocol.

Membrane Preparation. Transfected HEK cells were grown to 70 to 80% confluence and briefly washed with ice-cold HEPES-buffered saline/EDTA (2 × 10 ml; 10 mM HEPES, 0.9% NaCl, and 0.2% EDTA, pH 7.4). All of the subsequent steps were conducted at 0–4°C. Cells were lifted from the flask with HEPES-buffered saline/EDTA. A cell pellet was recovered by centrifugation (250g for 5 min) and the pellet homogenized using a Polytron (PT2100; Kinematica AG, Littau/Lucerne, Switzerland) (20,000 rpm, 5 × 10-s bursts) in 10 mM HEPES and 10 mM EDTA, pH 7.4. The homogenate was centrifuged (50,000g, 20 min) and the pellet rehomogenized and centrifuged as described previously in a low EDTA buffer (10 mM HEPES and 0.1 mM EDTA, pH 7.4). The final pellet was suspended in this buffer at a concentration of 4 to 5 mg of protein/ml and stored at –80°C until required.

[³H]NMS Binding. A range of concentrations of [³H]NMS was used (0.07–3 nM) to construct saturation binding curves. Nonspecific binding was determined in the presence of 1 μM atropine. Transfected HEK cell membranes (10–20 μg) were incubated at room temperature for 90 min in assay buffer (10 mM HEPES and 1 mM MgCl₂, pH 7.4). Reactions were terminated by rapid vacuum filtration over Whatman GF/B filters and four washes with ice-cold assay buffer, and radioactivity on the filter was quantified by scintillation counting. Agonist and antagonist affinities were determined using membrane preparations (10–20 μg) from HEK cells expressing wild-type (WT)-M₃ or ^N514YM₃. Competition binding was performed in the presence of approximately 0.4 nM [³H]NMS for 2 h at 37°C.

Receptor Up-Regulation Studies. HEK-293 cells transfected with either WT-M₃ or ^N514YM₃ receptor were incubated with the indicated concentrations of antagonist for 24 h. At the end of this period, cells were washed rapidly with 3 × 1 ml of Krebs-Henseleit buffer (KHB) (118.6 mM NaCl, 4.7 mM KCl, 1.3 mM CaCl₂, 10 mM HEPES, 11.7 mM d-glucose, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, and 25 mM NaHCO₃, pH 7.4) and then incubated for 60 min at 37°C in 500 μl of KHB with approximately 5 nM [³H]NMS in the absence or presence of 1 μM atropine. Free radioligand was removed after transfer to ice by rapid washing of wells with 3 × 1 ml of ice-cold KHB. Cell monolayers were dissolved in 250 μl of 0.1 M NaOH, and radioactivity was determined.

[³H]Inositol Phosphate Accumulation. HEK-293 cells transfected with either WT-M₃ or ^N514YM₃ receptor were grown to 60 to 70% confluence and then labeled with 1 μCi/ml of myo-[2-³H]inositol for 24 h. Cell monolayers were washed and incubated in KHB at 37°C. For agonist additions, cells were incubated with LiCl (10 mM) for 15 min before agonist addition for a further 10 min. For inverse agonist additions, cells were incubated with the inverse agonist for 15 min before the addition of LiCl (10 mM), and incubations were continued for 10 min. In either case, incubations were terminated by rapid aspiration of buffer and addition of ice-cold trichloroacetic acid (0.5 M). The [³H] total inositol (poly)phosphate (IP_x) fraction was recovered exactly as described previously (Batty et al., 1997).

Quantification of M₃ mACh Receptor Phosphorylation. HEK-293 cells transfected with either WT-M₃ or ^N514YM₃ cDNA were grown to 80 to 90% confluence in six-well plates. Receptor phosphorylation was determined as described by Tobin (1997). Briefly, cells were incubated in phosphate-free KHB buffer containing [³²P]orthophosphate (50 μCi/ml) for 90 min at 37°C. The cells were either incubated with inverse agonists throughout the preincubation period or for 10 min in experiments with agonists. Incubations were terminated by aspiration and addition of ice-cold solubilization buffer (10 mM Tris, 10 mM EDTA, 500 mM NaCl, 0.2% SDS, 1% NP-40, 0.5% sodium deoxycholate, 0.2 mM sodium vanadate, and 2 mM disodium nitrophenyl phosphate, pH 7.4) for 30 min. Samples were centrifuged (14,000 rpm, 5 min, 4°C) and the supernatant incubated with M₃ antibody on ice for 60 min. The antibody-receptor complex was recovered by incubation with protein A-Sepharose for 1 h at 4°C. The protein A-Sepharose-bound complex was recovered by centrifugation (14,000 rpm, 1 min), and the beads were washed four times in Tris/EDTA buffer (10 mM Tris/HCl, 10 mM EDTA, pH 7.4). The final pellet was resuspended in 2× SDS-polyacrylamide gel electrophoresis sample buffer (10 mM Tris/HCl, pH 6.8, 200 mM dithiothreitol, 4% SDS, 0.1% bromphenol blue, and 20% glycerol) and incubated for 2 min at 85°C. Samples, equalized for receptor number (by [³H]NMS binding), were loaded onto an 8% SDS-polyacrylamide gel electrophoresis gel and run at 180 mV for 30 to 60 min. Radioactivity was visualized by autoradiography and scintillation counting.

Data and Statistical Analysis. Data are shown as mean ± S.E.M. for the indicated number of experiments. Saturation binding data were fitted with hyperbolae (one-site binding) using GraphPad Prism 3.0 (GraphPad Software, San Diego, CA). B_max and K_D values were derived from these curves. Competition binding curves and functional concentration-response curves were fitted to a four-parameter logistic equation using GraphPad Prism 3.0. The best fit between a variable Hill coefficient and a Hill coefficient fixed to unity was determined using an F-test. IC₅₀ values, generated by these inhibition curves, were corrected to give binding constant (K_i) values for each test compound using the Cheng-Prusoff equation (Cheng and Prusoff, 1973). The statistical significance of differences between data was determined using either Student's t test, with K_i and K_B values first being converted to the respective normally distributed negative logarithm (pK_i or pK_B) or one-way analysis of variance with Dunnett's post test for multiple comparisons.

Results

Characterization of mACh Receptor Populations Using [³H]NMS Binding. The HEK-293 cell clone used in this study possesses a small (B_max 34 ± 2 fmol/mg protein), endogenous M₃ mACh receptor population. Transient transfection of HEK cells with either WT-M₃ or ^N514YM₃ mACh receptor cDNA (20 μg/175-cm² flask) resulted in a marked increase in specific [³H]NMS binding sites (Fig. 1). Despite equal plasmid loading, the WT-M₃ mACh receptor was expressed at consistently higher (∼3-fold) densities than the ^N514YM₃ mutant receptor (see Fig. 1C). In addition, the K_D determined for [³H]NMS binding was significantly (p < 0.05) greater at the ^N514YM₃ receptor (269 ± 10 pM) than at the WT-M₃ receptor (169 ± 7 pM). In future experiments, receptor expression was adjusted to give approximately equal numbers of WT-M₃ or ^N514YM₃ mACh receptors, and individual K_D values were determined in each experiment to allow accurate determination of binding parameters (e.g., conversion of IC₅₀ to K_i values).

Antagonist Affinity at Wild-Type and N514Y M₃ mACh Receptors. To examine whether the N514Y point mutation caused a consistent change in antagonist affinity, we assessed the ability of atropine, 4-DAMP, pirenzepine, and methoctramine to displace [³H]NMS binding (Fig. 2). Although all the antagonists tested showed small reductions in binding affinity for the ^N514YM₃ versus WT-M₃ mACh receptor, statistically significant reductions were only seen for atropine (pK_i: WT, 9.16 ± 0.09; N514Y, 8.85 ± 0.07) and 4-DAMP (pK_i: WT, 8.99 ± 0.03; N514Y, 8.60 ± 0.04; p < 0.05). Slope factors were not significantly different to unity, except for methoctramine (n_H: WT 1.56 ± 0.05; N514Y, 1.62 ± 0.10), which might be attributable to this agent binding to both orthosteric and allosteric binding sites and/or possessing slow dissociation kinetics at mACh receptors (Lee and El-Fakahany, 1991).

These data suggest that atropine and 4-DAMP are behaving as inverse agonists at the ^N514YM₃ mACh receptor. If the N514Y mutation has induced a degree of agonist-independent G protein coupling, then it may be possible to detect a greater difference in inverse agonist affinities between WT and ^N514YM₃ receptors if the proportion of uncoupled receptors is increased, provided there is sufficient G protein available in the system (Strange, 2002). This has been tested by supplementing the binding assay buffer with GTP (100 μM); under these conditions, no additional effects on pIC₅₀ values were observed for any of the inverse agonist/antagonists studied here (data not shown).

The effect of the N514Y mutation on agonist binding affinity was also assessed (Fig. 3). Both MCh and pilocarpine exhibited significantly higher apparent binding affinities at the ^N514YM₃ compared with the WT-M₃ mACh receptor (pK_B values: WT, 4.89 ± 0.07; N514Y, 6.01 ± 0.06 for MCh; p < 0.005; WT, 5.47 ± 0.04; N514Y, 6.16 ± 0.06 for pilocarpine; p < 0.005). Again, inclusion of GTP (100 μM) did not affect apparent affinity estimates for either MCh or pilocarpine (data not shown).

Effects of mACh Receptor Inverse Agonist/Antagonists on [³H]IP_x Accumulation. To examine the effects of receptor mutation on the ability to activate a G protein-dependent signaling output, [³H]IP_x accumulation in intact HEK cells was assessed. Transfection of cells with ^N514YM₃ mACh receptor cDNA (≥0.5 μg) resulted in a B_max value of 1140 ± 80 fmol/mg protein. In contrast, a higher maximal receptor expression level could be obtained for the WT-M₃ receptor (B_max 3210 ± 159 fmol/mg protein). Therefore, HEK cells were routinely transfected with 0.5 μg of N514Y mutant receptor cDNA or an amount of WT-M₃ cDNA that would result in a similar receptor expression level (typically 0.15 μg). In each experiment, the matching of mACh receptor expression levels was checked using [³H]NMS binding. The addition of Li⁺ (10 mM) to HEK cells expressing ^N514YM₃, but not WT-M₃, mACh receptors caused a near-linear increase in [³H]IP_x accumulation such that after only 10 min of Li⁺ addition, [³H]IP_x accumulation was approximately 300% higher in the ^N514YM₃ receptor-expressing cells. It should be noted that although stimulation of the small, endogenous M₃ mACh receptor population generated a measurable [³H]IP_x response in the presence of Li⁺, this was small (typically <25% of that seen in transfected cells). Nevertheless, empty vector-transfected HEK cells were always included to assess any potential influence of the endogenous receptor population.

To assess inverse agonism, HEK cells expressing either ^N514YM₃ or WT-M₃ mACh receptors were incubated with atropine, 4-DAMP, pirenzepine, or methoctramine for 15 min before Li⁺ addition for a further 10 min (Fig. 4). All the agents behaved as inverse agonists concentration-dependently decreasing [³H]IP_x accumulation in the presence of Li⁺ to levels observed in the absence of Li⁺. These data suggest that all four antagonists behave as “full inverse agonists” with respect to the M₃ mACh receptor coupling to phosphoinositide hydrolysis. Comparison of half-maximal inhibitory concentrations revealed that whereas pirenzepine and methoctramine exhibited no significant differences in their IC₅₀ values with respect to decreasing basal [³H]IP_x accumulation via WT-M₃ or ^N514YM₃ mACh receptors, atropine (pIC₅₀, WT, 8.84 ± 0.08; N514Y, 8.26 ± 0.02) and 4-DAMP (pIC₅₀, WT, 8.73 ± 0.02; N514Y, 8.39 ± 0.01) showed significant (3.8- and 2.2-fold, respectively) differences (p < 0.05).

The effects of the agonists MCh, oxotremorine, and pilocarpine were next assessed. Because receptor expression influences both receptor efficacy and potency, particular care was taken to ensure that ^N514YM₃ and WT-M₃ mACh receptor expression levels were matched. Each agonist stimulated a concentration-dependent increase in [³H]IP_x accumulation (Fig. 5) with all the agonists exhibiting 14- to 17-fold higher potency in HEK cells expressing the ^N514YM₃ versus WT-M₃ mACh receptor (pEC₅₀ values: MCh, 6.18 ± 0.06 versus 7.40 ± 0.06; oxotremorine, 6.49 ± 0.08 versus 7.67 ± 0.09; pilocarpine, 5.39 ± 0.04 versus 6.53 ± 0.03). The agonists also exhibited a spectrum of relative intrinsic activities. Thus, compared with MCh (1 mM), oxotremorine and pilocarpine exhibited intrinsic activities of 73 and 51% in WT-M₃ receptor-expressing HEK cells; these values increased to 97 and 63%, respectively, in cells expressing the ^N514YM₃ receptor (Fig. 5).

Effects of Chronic Exposure to Inverse Agonists on Receptor Expression Levels. Earlier experiments showed that at equivalent plasmid loading, the cell-surface expression of WT-M₃ receptor was approximately 3-fold higher than that of the ^N514YM₃ receptor (see Fig. 1). Therefore, we have assessed whether coincubation with agents shown to suppress ^N514YM₃ mACh receptor constitutive activity can affect receptor expression levels. In these experiments, HEK cells have been transfected with 0.2 μg of plasmid DNA/well for 24 h before inverse agonist addition. Incubation of ^N514YM₃ receptor-expressing HEK cells for 24 h with atropine, 4-DAMP, or pirenzepine caused concentration-dependent increases in receptor expression levels (Fig. 6A). It should be noted that meaningful data could not be obtained for methoctramine (probably because of the slow dissociation kinetics of this agent and a consequent inability to wash it out sufficiently to allow [³H]NMS occupation of all the cell-surface receptors). Atropine and pirenzepine each increased ^N514YM₃ mACh receptor expression levels to a similar level (93 ± 3 and 84 ± 2% increases, respectively) but with different potencies (pEC₅₀ values: atropine, 8.22 ± 0.11; pirenzepine, 6.33 ± 0.09). Although 4-DAMP behaved as a “full” inverse agonist with respect to suppressing basal [³H]IP_x accumulation, when assessed for receptor regulation, it was only approximately 50% as efficacious as atropine (maximum up-regulation 47 ± 4%, pEC₅₀, 7.81 ± 0.20).

The agonist-mediated receptor up-regulation of the ^N514YM₃ receptor by atropine, pirenzepine, and 4-DAMP appeared to occur linearly over the 24-h incubation period (Fig. 6A, inset). In addition, cell surface expression of either the endogenous M₃ mACh receptor population in HEK cells or the recombinant WT-M₃ receptor was unaffected by 24-h incubations with any of these agents (see Fig. 6B).

Effects of Inverse Agonists on Phosphorylation of the Wild-Type and N514Y Mutant M₃ mACh Receptor.³²P_i labeling of HEK cells has allowed basal and agonist-stimulated WT-M₃ and ^N514YM₃ mACh receptor phosphorylation to be assessed. Basal phosphorylation of the ^N514YM₃ receptor was 50 to 60% greater than for the WT-M₃ receptor (WT, 8304 ± 412; N514Y, 12,898 ± 496 absorbance units/mm²; p < 0.01; Fig. 7). Despite the increased basal receptor phosphorylation, MCh (1 mM) caused similar maximal increases in phosphorylation of the WT-M3 and ^N514YM₃ mACh receptors (Fig. 7B). Inclusion of atropine (1 μM present throughout the ³²P_i labeling period) had no effect on WT-M₃ receptor phosphorylation but significantly reduced phosphorylation of the ^N514YM₃ mACh receptor phosphorylation (by 56 ± 6%; p < 0.01; see Fig. 7B). These data indicate that the ^N514YM₃ mACh receptor exhibits constitutive phosphorylation that can be suppressed to WT-M₃ receptor basal phosphorylation levels by an inverse agonist.

Like atropine, 4-DAMP (1 μM) and pirenzepine (10 μM) also significantly reduced the constitutive phosphorylation of the ^N514YM₃ mACh receptor (Fig. 8). In contrast, methoctramine (100 μM), a full inverse agonist with respect to constitutive [³H]IP_x accumulation, failed significantly to affect constitutive ^N514YM₃ receptor phosphorylation (Fig. 8). Unfortunately, the technically demanding nature of these experiments precluded concentration dependencies for the effects of the inverse agonists being established.

Discussion

Constitutively active mutant (CAM) M₁–M₅ mACh receptors have previously been created by the introduction of a double point mutation at the junction of TM6 with the third extracellular loop (Spalding et al., 1997; Ford et al., 2002), corresponding to an N514Y, T515P mutation of the human M₃ mACh receptor. Huang et al. (1999) have reported that it is the homologous S388Y (and not the T389P) mutation of the human M₁ mACh receptor that is primarily responsible for the constitutive activity of the mutant receptor. Here, we show that introduction of the N514Y mutation into the human M₃ mACh receptor alone is sufficient to induce a robust level of constitutive activity and have pharmacologically compared WT-M₃ and ^N514YM₃ mACh receptors with respect to multiple signaling outputs.

Antagonist binding profiles showed that only mACh antagonists with relatively high affinity for the WT-M₃ receptor exhibited a significant decrease in binding affinity for the mutant ^N514YM₃ receptor. A decrease in inverse agonist affinity is predicted if it is assumed that the N514Y mutation induces a conformation that (at least partially) resembles the active state of the receptor and that the inverse agonist preferentially binds to the inactive (R) state of the receptor (de Ligt et al., 2000; Strange, 2002). Compared with the increases in agonist affinity observed for the mutant receptor, the reductions in antagonist affinity are modest. However, it has been shown that a 2-fold reduction in inverse agonist apparent affinity is predicted for a constitutively active mutant in which 50% of receptors are in the active (R*) state (relative to no detectable constitutive activity in the WT receptor) (Wade et al., 2001). The lack of affinity differences seen for pirenzepine and methoctramine may indicate that these compounds exert their inverse agonism via a mechanism distinct from a stabilization of R. Support for this idea comes from the observation that although all the inverse agonists examined here inhibited agonist-independent [³H]IP_x accumulation to similar extents, pEC₅₀ values for the ^N514YM₃ were significantly reduced, relative to the WT-M₃ receptor, for atropine and 4-DAMP but not for pirenzepine or methoctramine.

One mode of inverse agonist action, which allows an inverse agonist to suppress basal activity, but not exhibit a decreased affinity at a constitutively active mutant receptor, has recently been proposed (Strange, 2002). This does not involve a redistribution of receptor states but one in which the inverse agonist switches the active conformation of the receptor to an inactive state that still retains the ability to bind and sequester G proteins but is unable to activate them (McLoughlin and Strange, 2000; Strange, 2002; Monczor et al., 2003). A species of receptor that is inactive but still binds G protein is one of the key features of the cubic ternary complex model (Weiss et al., 1996) and one that distinguishes it from the extended ternary complex model.

At equivalent plasmid loading, the cell-surface expression level of the ^N514YM₃ mACh receptor was approximately 3-fold lower than that for the WT-M₃ receptor. The ^N514YM₃ (but not the WT-M₃) receptor expression level could be significantly increased by chronic (24 h) incubation in the presence of an inverse agonist. Ligand-mediated receptor up-regulation has been reported previously for constitutively active mutant receptors (Smit et al., 1996; Stevens et al., 2000; Li et al., 2001; Pauwels and Tardif, 2002), although the precise molecular mechanisms involved are still debated (Parnot et al., 2002). For some CAM-GPCR, the degree of up-regulation is positively correlated with both the degree of constitutive activity in the system and the ability of a ligand to decrease basal activity. The up-regulation of a CAM H₂ histamine receptor has been shown by preincubation with ligands, such as cimetidine and ranitidine, which were characterized as inverse agonists by their ability to decrease basal and forskolin-induced cyclic AMP production. Burimamide, which behaved as a neutral antagonist in cyclic AMP assays, did not affect the level of the H₂ histamine receptor expression (Smit et al., 1996). This suggests that the receptor may be down-regulated as a consequence of productive G protein coupling. If a CAM-GPCR functions in a similar way to its cognate agonist-induced WT counterpart, it would follow that only inverse agonists would cause up-regulation by virtue of stabilizing the inactive conformation (Parnot et al., 2002). However, other studies (Gether et al., 1997; Pauwels and Tardif, 2002) have shown that the creation of a CAM can diminish stabilizing constraints within the receptor, leading to an inherently unstable receptor that is more susceptible to destabilization and/or proteolytic degradation (Stevens et al., 2000). In this case, the expression level of the mutant is increased by any ligand [(inverse) agonist/antagonist] regardless of its efficacy.

In this study, three of the antagonists, classified as inverse agonists from their ability to decrease basal [³H]IP_x accumulation, up-regulated the expression of the ^N514YM₃ receptor in a time- and concentration-dependent manner. This effect was specific to the ^N514YM₃ receptor because no significant effect on the receptor density could be detected for either the recombinant WT-M₃ or endogenous HEK-M₃ receptor population. Despite being a full inverse agonist in [³H]IP_x assays, 4-DAMP was only 50% as efficacious as atropine with respect to ^N514YM₃ receptor up-regulation. It is noteworthy that 4-DAMP possesses a permanent positive charge, which is likely to reduce its membrane permeability. It has been shown that membrane-permeable ligands are able to facilitate the endoplasmic reticulum export of the δ-opioid receptor to the cell surface by acting as molecular “chaperones” (Petaja-Repo et al., 2002). Thus, up-regulation of a CAM μ-opioid receptor by naloxone methiodide (which, like 4-DAMP, possesses a permanent positive charge) was only 50% as efficacious as naloxone (Li et al., 2001). It is possible that in the ^N514YM₃-HEK system inverse agonists increase cell surface receptor expression by increasing both receptor stability and facilitating export of the CAM receptor through its intracellular processing pathway. Therefore, the lower efficacy of 4-DAMP may result from its inability to cross the lipid membrane and to facilitate this process. Up-regulation, by both antagonists and agonists of a constitutively active chimeric M₃ mACh receptor, has also been reported (Zeng et al., 2003). This study showed that inverse agonist-mediated inhibition of proteosomal degradation of the CAM receptor is one potential mechanism involved in up-regulation. Interestingly, this study also showed a correlation between ligand binding affinity and receptor up-regulation for all the compounds tested, except 4-DAMP. The authors suggested that one possible reason for this discrepancy was that 4-DAMP is unstable in aqueous solution. However, we have found no decrease in 4-DAMP apparent binding affinity under the conditions used here (data not shown); therefore, the reason for the behavior of 4-DAMP remains unclear.

The M₃ mACh receptor has been shown to be phosphorylated in response to agonist stimulation by a variety of protein kinases, including GPCR kinases (Willets et al., 2003) The observation that the ^N514YM₃ mutant is significantly more phosphorylated than the WT-M₃ suggests that the CAM receptor may adopt a conformation more closely resembling that of the agonist-bound WT-M₃ receptor. The human M₃ receptor possesses a large number of potential serine/threonine phospho-acceptor sites (>50 in the i3 loop alone), and we presently have no information on whether the increased basal phosphorylation of ^N514M₃ is at similar or different sites to those modified during agonist-dependent phosphorylation.

The increased basal phosphorylation of the ^N514YM₃ mACh receptor raises the intriguing possibility that this mutant receptor is also constitutively desensitized, and consequently, the observed degree of constitutive activity may well be underestimated. Quantification of the extent to which atropine, 4-DAMP, pirenzepine, and methoctramine decrease basal ^N514YM₃ phosphorylation also revealed differences with respect to the behaviors of these inverse agonists as suppressors of [³H]IP_x accumulation. Thus, methoctramine, which was a full inverse agonist in the latter assay, failed significantly to attenuate basal phosphorylation of the ^N514YM₃ receptor. Recent evidence obtained for the AT_1A (Thomas et al., 2000) and μ-opioid (Yu et al., 1997) receptors has shown that G protein coupling and receptor phosphorylation and internalization are not necessarily a sequential series of events and that some agonists are able to induce receptor phosphorylation and/or internalization without detectable G protein coupling. “Ensemble theory” (Kenakin and Onaran, 2002) suggests that the behavior of each GPCR involves the stabilization of a microconformation and that the unique selective affinity of a ligand changes the redistribution of receptor conformations to initiate or inhibit a response. Therefore, it is possible that methoctramine stabilizes conformation(s) of the ^N514YM₃ receptor that inhibit G protein coupling but are still constitutively phosphorylated.

In summary, the ^N514YM₃ mACh receptor is an interesting CAM for pharmacological analysis because it displays classic hallmarks of constitutive activity, including 1) increased agonist-independent PLC activity, which is concentration-dependently inhibited by inverse agonists, 2) decreased receptor stability/expression that can be rescued by inverse agonists, 3) increased agonist efficacy and affinity and decreased antagonist affinity, and 4) increased levels of “basal” receptor phosphorylation. Although all the antagonists studied here can be classified as full inverse agonists based on their ability to decrease agonist-independent [³H]IP_x accumulation, we have found interesting differences in the pharmacology of these compounds when receptor expression levels and receptor phosphorylation responses were examined.