Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Muscarinic acetylcholine (ACh) receptors (mAChRs; M₁-M₅; Caulfield and Birdsall, 1998) are members of the G protein-coupled receptor family and represent important targets for drug design and development. Functionally, M₁, M₃, and M₅ subtypes couple to the activation of phospholipase C (PLC) through the pertussis toxin-insensitive G_q/11 family of G proteins; and M₂ and M₄ subtypes couple to the inhibition of adenylyl cyclase through the pertussis toxin-sensitive G_i/o family of G proteins (Hulme et al., 1990; Caulfield, 1993). Previous molecular modeling studies (Ward et al., 1992; Nordvall and Hacksell, 1993) and site-directed mutagenesis and pharmacological studies (Fraser et al., 1989; Wess et al., 1991, 1992; Blüml et al., 1994; Huang et al., 1999) have identified several highly conserved residues critical for agonist binding and receptor activation; however, the molecular mechanisms by which receptors are activated on acetylcholine (ACh) binding are still not clear.

A mutant human M [HM₅(S465Y, T466P)] receptor showed significant constitutive activity, increased agonist potency, and binding affinity (Spalding et al., 1995). The HM₅ Ser465 is conserved in M₁ receptors and is an Asn residue in M₂, M₃, and M₄ receptor subtypes; whereas the HM₅ Thr466 is conserved in all five subtypes of mAChRs (Ser388 and Thr389 in HM₁ receptors; see Fig. 1). In previous studies (Huang et al., 1998), we demonstrated that an equivalent mutant HM₁ receptor [HM₁(S388Y, T389P), a mutant HM₁ receptor with Ser388 replaced by Tyr and Thr389 replaced by Pro], stably expressed in A9 L cells, showed significantly enhanced agonist potency, binding affinity, and G protein coupling. The enhancement is neither expression level nor cell line dependent but rather is an intrinsic property of the mutant receptor (X.-P. H., F. E. W., S. M. P. and W. S. M. O., submitted for publication). In contrast to the high level of constitutive activity observed for HM₅(S465Y, T466P) receptors, HM₁(S388Y, T389P) receptors showed limited constitutive activity (~20%) at high expression levels, but not at low expression levels ( X.-P. H., F. E. W., S. M. P. and W. S. M. O., submitted for publication). However, HM₁(S388Y, T389P) receptors can be activated by choline, which is found in Dulbecco's modified Eagle's medium, and unknown agonists in FBS (Huang et al., 1998). The differences observed between mutant HM₁ and HM₅ receptors might be due to several factors, such as subtype (HM₁ versus HM₅ receptors), cell line (A9 L versus NIH 3T3 cells), receptor functional assay [phosphatidylinositol (PI) hydrolysis assay versus receptor selection and amplification technology (R-SAT)], expression mode (stable versus transient expression), or expression level. However, the common features, enhanced agonist potency and binding affinity, observed at HM₅(S465Y, T466P) and HM₁(S388Y, T389P) receptors suggest that the junction of transmembrane domain (TM) VI and the N-terminal of the third extracellular loop (Ne3) has a conserved functional role in mAChRs. This is supported by preliminary results indicating that equivalent mutations (AsnThr-to-TyrPro mutations) in M₂ and M₃ receptors also result in increased agonist binding affinity (Ellis et al., 1998).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   A model of the HM1 receptor amino acid sequence depicting seven TMs. Residues involved in agonist binding and receptor activation are highlighted in gray, including Asp105 in TM III, Thr192 in TM V, and Tyr381 and Asn382 in TM VI. The Ser388 and Thr389 residues are circled. The equivalent residues in other subtypes of mAChRs are listed for comparison.

It is widely accepted that ACh binds to a binding cavity, about two helical turns below the membrane surface, formed within the TM domains (Hulme et al., 1990; Wess 1993, 1996). Ser388 and Thr389 are located at the junction of TM VI and Ne3 and about two helical turns above the Tyr381 and Asn382 residues in TM VI (see Fig. 1), which are critical residues involved in ACh binding and/or receptor activation (Ward and Hulme, 1997; Huang et al., 1999). Therefore, Ser388 and Thr389 probably are not the primary ligand binding sites. In fact, several single mutations at Ser465 of HM₅ receptors cause varying degrees of constitutive activity (measured by R-SAT) in which basic and bulky substitutions are more effective than acidic and small substitutions (Spalding et al., 1997). The mutations at Ser465 or Ser465 and Thr466 appear to cause the formation of active receptor states (R*; Spalding et al., 1995, 1997) in accordance with the allosteric ternary complex model for G protein-coupled receptors (Samama et al., 1993). Mutations of Ser388 and Thr389 to Tyr and Pro, respectively, in HM₁ receptors may induce the mutant receptor to form structurally and conformationally flexible states that are favorable for agonist binding and activation (Huang et al., 1998), as suggested recently in mutant ₂-adrenergic receptors (Kobilka et al., 1998).

To determine the functional role of individual substitutions in HM₁(S388Y, T389P) receptors, we created A9 L cell lines stably expressing HM₁(S388Y) or HM₁(T389P) receptors and characterized the mutant receptors. Here we present results indicating that the Ser388Tyr mutation is responsible for enhanced agonist potency and agonist binding affinity, whereas the Thr389Pro mutation is responsible for altering G protein interactions.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials and methods used in this study were as reported previously (Huang et al., 1998, 1999) or described below.

Muscarinic Agonists, Chemicals, and Other Materials. Muscarinic agonists bethanechol and methacholine were purchased from Research Biochemical Inc. (Natick, MA). GTP sodium salt was ordered from Sigma Chemical Co. (St. Louis, MO). Filtermate 196, UniFilter GF/B, MicroScint 20, TopSeal A, and Backing Tape were purchased from Packard Instrument Company (Meriden, CT). Deep-well microtiter plates (96-well and 1.2 ml/well) were obtained from Marsh Biomedical Products, Inc. (Rochester, NY).

Mutation Strategy. Mutations of Ser388 to Tyr or Thr389 to Pro were carried out using the QuickChange kit (Stratagene, La Jolla, CA) using HM₁pcD1 provided by Dr. Tom I. Bonner (Bonner et al., 1988). All primers used in mutations were synthesized by Life Technologies (Grand Island, NY). The sense primer for the Ser388Tyr mutation was 5'-GGTGCTGGTGTACACCTTCTGCAAGG-3' (with the changed base in bold), and the antisense primer was 5'-CCTTGCAGAAGGTGTACACCAGCACC-3'. The sense primer for the Thr389Pro mutation was 5'-GGTGCTGGTGTCCCCCTTCTGCAAGG-3' (with the mutated base in bold), and the antisense primer was 5'-CCTTGCAGAAGGGGGACACCAGCACC-3'. The mutations were confirmed by dideoxy nucleotide sequencing using the T7 Sequenase sequencing kit from Amersham Life Science (Arlington Heights, IL).

Creation of Stable A9 L Cell Lines. A9 L cells (American Type Culture Collection, Rockville, MD) were cotransfected with HM₁(S388Y)pcD1 or HM₁(T389P)pcD1 and pNEOGAL (Stratagene) according to the calcium phosphate method (Chen and Okayama, 1987). Transfected A9 L cells were selected in the presence of 800 µg/ml G418 (Fisher, Pittsburgh, PA) and screened as described previously (Huang et al., 1998). A9 L cells stably expressing HM₁(S388Y) or HM₁(T389P) receptors were subcultured for functional and binding studies.

Radioligand Binding Assays-TopCount NXT System. [³H](R)-3-Quinuclidinyl benzilate [(R)-QNB] saturation binding assays and ligand inhibition binding assays were performed as described previously (Huang et al., 1998), except that deep-well microtiter plates (96-well) were used instead of glass tubes. Reactions were initiated by the addition of membrane proteins to mixtures of reagents. The plates were sealed with Parafilm and incubated at room temperature for 2 h. The incubation was terminated by transfer to a 96-well UniFilter (GF/B) using Filtermate 196. The UniFilter was washed 4 times with 1 ml of cold binding buffer (25 mM sodium phosphate, pH 7.4, containing 5 mM magnesium chloride). The UniFilter then was dried in a fume hood for at least 1 h, and its back was sealed with Backing Tape. To each well was added 50 µl of MicroScint 20, and the top of the plate was sealed with TopSeal A. The filter was soaked overnight, and the radioactivity was counted in the TopCount NXT system (Packard) running Microsoft Windows NT 4.0. There were no differences between results obtained from the TopCount NXT system and a traditional liquid scintillation count system (data not shown).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Receptor Expression and Antagonist Binding Properties. HM₁(S388Y) and HM₁(T389P) receptors were stably expressed in A9 L cells at levels of 2600 ± 160 (mean ± S.E.M.) and 1200 ± 210 fmol/mg, respectively. They both showed a small (1.8- to 3.8-fold) but significant (P < .05) reduction in binding affinity for [³H](R)-QNB compared with HM₁[wild-type (WT)] and/or HM₁(S388Y, T389P) receptors. HM₁(T389P) receptors also showed a 1.7-fold lower (P < .01) binding affinity for [³H](R)-QNB than HM₁(S388Y) receptors (Table 1). When other classic muscarinic antagonists were examined, HM₁(S388Y) receptors showed the same antagonist binding profiles as HM₁(WT) and/or HM₁(S388Y, T389P) receptors (P > .05). In contrast, HM₁(T389P) receptors displayed varying degrees of reduced affinity for these antagonists (2.8- to 16-fold; P < .05) compared with HM₁(WT) and/or HM₁(S388YY, T389P) receptors.

Pharmacology of HM₁(S388Y) and HM₁(T389P) Receptors. Muscarinic agonists used to characterize HM₁(S388Y, T389P) receptors (Huang et al., 1998) were also examined at HM₁(S388Y) and HM₁(T389P) receptors to compare maximal responses and potencies in stimulating PI hydrolysis. All PI hydrolysis experiments were conducted in Krebs-Henseleit buffer instead of media to exclude the effects of choline and other potential muscarinic agonists in animal serum (Huang et al., 1998).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Pharmacology of muscarinic agonists at HM1(S388Y) and HM1(T389P) receptors. Data for HM1(WT) and/or HM1(S388Y, T389P) receptors are from Huang et al. (1998) and are included for comparison. PI hydrolysis assays were conducted in Krebs-Henseleit buffer. Data represent the mean ± S.E. from a minimum of three assays, each performed in duplicate. A, ACh. B, CCh. C, Oxo-M. D, Oxo. E, arecoline. F, choline.

Agonist Binding Properties at HM₁(S388Y) and HM₁(T389P) Receptors. Consistent with enhanced agonist activity at HM₁(S388Y) receptors, muscarinic agonists also displayed significantly increased binding affinities at HM₁(S388Y) eceptors compared with HM₁(WT) receptors (Huang et al., 1998, 1999) or HM₁(T389P) receptors (Table 3 and Fig. 3). The smallest changes were observed for choline, which showed essentially the same binding affinity for HM₁(S388Y) receptors as HM₁(WT) receptors, yet 2.7-fold higher affinity than for HM₁(T389P) receptors. Except for Oxo-M, which showed more than 15-fold higher binding affinity for HM₁(S388Y) receptors than for HM₁(S388Y, T389P) receptors (Huang et al., 1998), all other agonists (ACh, CCh, Oxo-M, arecoline, and choline) showed similar binding affinity at HM₁(S388Y) receptors and HM₁(S388Y, T389P) receptors, and the difference in binding affinity (one-site binding model) was within 3.5-fold.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Muscarinic agonist binding properties at HM1(S388Y) and HM1(T389P) receptors. Agonist binding data (except Oxo-M) for HM1(WT) and HM1(S388Y, T389P) receptors are from Huang et al. (1998) and are included for comparison. Inhibition binding assays were conducted on membrane homogenates in the presence of 0.1 nM [3H](R)-QNB for HM1(S388Y) receptors and 0.3 nM [3H](R)-QNB for HM1(T389P) receptors. Data represent the mean ± S.E.M. from a minimum of three assays, each performed in triplicate. A, ACh binding in the absence and presence of Gpp(NH)p (100 µM). B, CCh binding in the absence and presence of GTP (400 µM). C, Oxo-M in the absence of GTP analogs. D, arecoline binding in the absence and presence of GTP (400 µM).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Mutations of Ser388Thr389 to TyrPro produced enhancement of agonist potency, binding affinity, and G protein coupling (Huang et al., 1998); however, because both residues were replaced concurrently, studies of the double mutations could not identify the relative roles of individual residues. In this study, we characterized two mutant receptors with single substitutions: HM₁(S388Y) and HM₁(T389P) receptors. Mutation of either Ser388 or Thr389 did not change the overall structure of the mutant receptors, as indicated by generally similar antagonist binding profiles for HM₁(S388Y) and HM₁(T389P) receptors as found previously for HM₁(S388Y, T389P) receptors and HM₁(WT) receptors. The greatest reduction was observed for NMS, with a 16-fold lower binding affinity for HM₁(T389P) than for HM₁(WT) receptors, suggesting that the binding pocket for NMS may extend to the Thr389 region.

In general, HM₁(S388Y) receptors functioned much like HM₁(S388Y, T389P) receptors, whereas HM₁(T389P) receptors more closely resembled HM₁(WT) receptors in PI hydrolysis assays. Agonists were much more potent at HM₁(S388Y) receptors than at HM₁(T389P) receptors. Antagonists slightly inhibited basal PI hydrolysis at HM₁(S388Y) receptors but not at HM₁(T389P) receptors. Consistent with the functional similarity between HM₁(S388Y) and HM₁(S388Y, T389P) receptors or between HM₁(T389P) and HM₁(WT) receptors, HM₁(S388Y) receptors had dramatically enhanced agonist binding affinity that resembled HM₁(S388Y, T389P) receptors, whereas HM₁(T389P) receptors showed similar agonist binding affinity to HM₁(WT) receptors. These data indicate that greatly enhanced agonist binding affinity may be a major contributor to enhanced agonist potency in HM₁(S388Y) receptors. In addition, the potency difference varied widely for different agonists and was not restricted to agonists with permanent positive charges in the amine head group. Agonists without hydrophobic side chains (e.g., CCh and choline) displayed lower increases in potency at HM₁(S388Y) receptors compared with WT receptors.

Bulky or basic substitutions (such as Phe, Arg, or Lys) at Ser465 of HM₅ receptors favor the formation of active receptor states leading to significant increases in agonist potency (measured by R-SAT) and high levels of constitutive activity (Spalding et al., 1997). However, the single Ser465Tyr mutation in HM₅ receptors was not identified or characterized. In the present study, the Ser388Tyr mutation did not result in a high level of constitutive activity in PI hydrolysis assays but rather produced significant increases in agonist potency and binding affinity. These data indicate that HM₁(S388Y) receptors are probably not in an active conformational state but in a state favorable for agonist binding and activation as suggested recently for the mutant ₂-adrenergic receptor (Kobilka et al., 1998).

HM₁(S388Y) receptors had enhanced agonist binding affinity similar to that of HM₁(S388Y, T389P) receptors but without a GTP-sensitive high-affinity binding site. For example, ACh interacted with three sites at HM₁(S388Y, T389P) receptors with the highest affinity binding sensitive to Gpp(NH)p, yet bound to two sites at HM₁(S388Y) receptors, with the high-affinity site insensitive to GTP. Choline and Oxo interacted with multiple sites at HM₁(S388Y, T389P) receptors but with only a single site at HM₁(S388Y) receptors. In contrast to HM₁(S388Y) receptors, HM₁(T389P) receptors had similar agonist binding affinities as HM₁(WT) receptors but gained an extra GTP-sensitive high-affinity binding site. For example, Oxo and arecoline bound to a single site at HM₁(WT) receptors, whereas two sites were observed for all tested agonists at HM₁(S388Y, T389P) receptors (Huang et al., 1998). In contrast, CCh, Oxo-M, and Oxo displayed three sites at HM₁(T389P) receptors, as did ACh at HM₁(WT) and HM₁(S388Y, T389P) receptors. The high-affinity binding sites of ACh, CCh, and arecoline were shifted or abolished by GTP or Gpp(NH)p, indicating that the extra high-affinity binding sites on HM₁(T3989P) receptors were associated with G protein interactions. The underlying molecular mechanisms of the modulation of receptor-G protein coupling are not clear at this point. Because the mutation was at the extracellular face of TM VI, it is possible that the mutation might produce conformational changes at the cytoplasmic side of TM VI, which is a critical determinant for G protein coupling (Wess, 1996). This is consistent with the recent finding that the third extracellular loop of the ₂-adrenergic receptor can modulate receptor-G protein interactions (Zhao et al., 1998).

HM₁(S388Y) receptors lost a GTP-sensitive high-affinity site compared with HM₁(S388Y, T389P) receptors, but muscarinic agonists showed similar maximal responses and potencies at these mutant receptors. In contrast, HM₁(T389P) receptors gained an extra GTP-sensitive high-affinity site and were expressed at a higher level compared with HM₁(WT) receptors, yet most agonists had similar activities. In fact, arecoline and Oxo-M displayed reduced potencies by 3.6- and 4.9-fold, respectively, at HM₁(T389P) receptors compared with HM₁(WT) receptors. Therefore, the changes in G protein coupling with HM₁(S388Y) or HM₁(T389P) receptors apparently were not associated with functional changes in PI hydrolysis. These data suggest that the effects of mutations on G protein coupling and receptor activity may be independent. The detected changes in G protein coupling may not reflect association with G_q/11 proteins but with other G proteins. Further investigations are necessary to address the molecular mechanisms underlying the changes in G protein coupling of the mutant receptors and the possibility that other G proteins are involved in coupling with the mutant receptors.

These data indicate a potential role for the junction of TM VI and Ne3 consistent with the proposed switch function of TM VI in receptor activation processes (Spalding et al., 1998). In addition, Ser388Thr389 may harbor allosteric binding sites that regulate the binding of ligands at the primary site, although HM₁(T389A) receptors did not exhibit changes in binding affinities for NMS, ACh, or an allosteric ligand, gallamine (Matsui et al., 1995). The Thr389 residue is highly conserved in mAChRs, and it is expected that equivalent mutations would cause similar effects in other members of the mAChR family.

The importance of the region in ligand binding and receptor function is not restricted to the mAChR family. A similar critical involvement of residues at equivalent positions in ligand binding and/or receptor function has been reported in many other G protein-coupled receptors [sequence alignment information is available from the database of mutants of family A G protein-coupled receptor (http://www-grap. fagmed.uit.no/GRAP/homepage.html; Kristiansen et al., 1996; Edvardsen and Kristiansen, 1997)]. For example, Glu297 in -opioid receptors (Hjorth et al., 1995; Jones et al., 1998) and the corresponding Trp284 in -opioid receptors (Valiquette et al., 1996) appear equivalent to Ser388 and are critical for the selectivity of opioid ligands. Asp268 in the B₂ bradykinin receptors (Kyle et al., 1994; Nardone and Hogan, 1994; Novotny et al., 1994) and the corresponding Gly273 in NK₂ receptors (Bhogal et al., 1994; also equivalent to Ser388), and Asp263Val264 of AT₁ receptors (Hjorth et al., 1994) and the corresponding Phe286Asp287 in human Y₁ neuropeptide Y receptors (Walker et al., 1994; Sautel et al., 1995) are important for agonist binding. The naturally occurring mutation of Ala593Pro in the human luteinizing hormone receptor (Ala593 corresponding to Thr389) does not change hormone binding affinity but abolishes G_s coupling (Kremer et al., 1995). The Tyr272 in NK₁ receptors (equivalent to Thr389) is also important in the selective binding of nonpeptide antagonists (Gether et al., 1993, 1994; Huang et al., 1994), and mutations of Tyr272 did not affect substance P binding but did decrease the Hill slope (Gether et al., 1994). In addition, mutations at Pro271 (equivalent to Ser388) in NK₁ receptors increased substance P binding affinity and decreased the Hill slope (Gether et al., 1994), suggesting that mutations in this region improve G protein coupling, in a similar fashion to the Ser388Tyr mutation in HM₁ receptors.

Taken together, the data indicate that the Ser388Tyr mutation is the major source of enhanced agonist potency and binding affinity, whereas the Thr389Pro mutation more subtly modified receptor interactions with G proteins. Enhanced agonist binding affinity, but not G protein coupling, appears to be fully responsible for the increased agonist potency observed at HM₁(S388Y) receptors. Ser388 and Thr389 are located about two helical turns above Tyr381 and Asn382, which are important for ACh binding and receptor function. This study demonstrates that Ser388 and Thr389 are potential targets for indirect modulation of agonist binding and G protein coupling, respectively. A combination of medicinal chemistry and pharmacological approaches could help identify ligands that can bind and induce receptor conformational changes to enhance endogenous ACh activity. This type of ligand might prove useful as a lead compound in the development of new treatments for neurological disorders such as Alzheimer's disease.