Abstract
Immunoprecipitation of a fusion protein between the α1b-adrenoceptor and Gα11 following a [35S]GTPγS [guanosine-5′-O-(3-thio)triphosphate] binding assay resulted in incorporation of low levels of nucleotide. The agonist phenylephrine increased incorporation some 30-fold. Agonist-induced binding represented 1.0 mol of [35S]GTPγS/mol of fusion protein. This was to the G protein linked to the receptor rather than endogenous Gαq/Gα11 as a fusion protein containing the α1b-adrenoceptor and a form of Gα11 (G208A) unable to exchange guanine nucleotides effectively, bound [35S]GTPγS very poorly. Fusion proteins between A293E, D142A, and 3CAM mutants of the α1b-adrenoceptor and Gα11bound substantially greater levels of [35S]GTPγS in the absence of agonist than the fusion incorporating the wild-type receptor. Constitutive binding of the nucleotide induced by these mutants was only 20% of the level achieved by phenylephrine. These mutant receptors thus do not provide an accurate mimic of the agonist-occupied state. Phentolamine reduced the binding of [35S]GTPγS and acted as a partial inverse agonist for each of the constitutively active mutants. [35S]GTPγS binding to Gα11 was elevated by phenylephrine in both wild-type and constitutively active mutant forms of the fusion proteins, but agonist potency and binding affinity were 50 times higher for the fusions containing the mutated receptors. These studies provide the first direct demonstration of the capacity of constitutively active mutants of a receptor to stimulate guanine nucleotide exchange on the α subunit of a Gq family G protein and defines a strategy potentially suitable for any receptor that couples to these G proteins.
Exchange of GTP for GDP on the α subunit of a heterotrimeric G protein represents the initial activation point for signal transduction mediated via G protein-coupled receptors (GPCRs). It is thus the earliest and most appropriate point to measure the effectiveness of GPCR-G protein interactions (Gierschik et al., 1994; Wieland and Jakobs, 1994). Despite the widespread use of ligand-regulated [35S]GTPγS binding assays to monitor these events, it has historically been extremely difficult to observe significant elevations of guanine nucleotide exchange produced by GPCRs that couple to G proteins other than members of the pertussis toxin-sensitive Gi family. At least in part, this reflects a combination of the low-basal guanine nucleotide exchange properties of Gq and the Gsfamily G proteins and the widespread expression profile of members of the Gi family.
Agonist-independent activation of signaling cascades by GPCRs has attracted great interest in the recent past (Scheer and Cotecchia, 1997; Leurs et al., 1998; Pauwels and Wurch, 1998; de Ligt et al., 2000). It is well established that mutations at a considerable range of positions in GPCRs can uncover or enhance such constitutive activity. The α1b-adrenoceptor has been particularly well studied in this regard (Allen et al., 1991; Kjelsberg et al., 1992;Perez et al., 1996; Scheer et al., 1996, 1997, 2000; Hwa et al., 1997;Lee et al., 1997; Mhaouty-Kodja et al., 1999; Rossier et al., 1999;McWhinney et al., 2000; Stevens et al., 2000). This reflects both that it was the first GPCR in which mutation was observed to generate such a constitutively active phenotype (Allen et al., 1991; Kjelsberg et al., 1992) and because mutations at a range of distinct locations produce such effects. However, although it is well appreciated that different mutations may produce varying levels of constitutive activity (Kjelsberg et al., 1992; Scheer et al., 1997; Stevens et al., 2000), this has been difficult to quantitate effectively. This reflects that this GPCR is coupled predominantly to the elevation of intracellular Ca2+ via Gq family G proteins. Measurements of functionality have had to be made, until now, either at the level of elevation of inositol phosphate production or via reporter gene assays. Because both of these are downstream endpoints of the initial G protein activation, they are subject to amplification and regulation that can influence details of pharmacology. Furthermore, mutational alterations can greatly alter the levels of expression of the α1b-adrenoceptor (Lee et al., 1997; Stevens et al., 2000), and changes in both the absolute levels of expression of a receptor and the stoichiometry of receptor to G protein are expected to modulate measured levels of constitutive activity.
By using fusion proteins (Seifert et al., 1999; Milligan, 2000) between forms of the α1b-adrenoceptor and the α subunit of the G protein G11, we now combine the defined GPCR-G protein stoichiometry of such constructs with their effective immunoprecipitation to directly monitor constitutive activity and ligand regulation of [35S]GTPγS binding to Gα11 produced by the wild-type and various constitutively active mutant (CAM) forms of the α1b-adrenoceptor.
Materials and Methods
Materials.
All materials for tissue culture were supplied by Invitrogen (Paisley, Strathclyde, UK). [3H]Prazosin (80 Ci/mmol) and [35S]GTPγS (1250 Ci/mmol) were from PerkinElmer Life Sciences (Boston, MA). Oligonucleotides were purchased from Cruachem (Glasgow, Strathclyde, UK). Receptor ligands were purchased from Sigma/RBI (Gillingham, Kent, UK). Production and characterization of the anti-Gq/G11 antiserum CQ was described by Mitchell et al. (1993). All other chemicals were from Sigma-Aldrich (Poole, Dorset, UK) and were of the highest grade available.
Construction of Fusion Proteins.
Production and subcloning of wild-type and mutated α1b-adrenoceptor-Gα11fusion proteins was performed in two separate stages. In the first step, the coding sequence of Gα11 was modified by PCR amplification using the amino terminal primer 5′GAGGACGGTACCACTCTGGAGTCCATG-3′; the initiating Met of Gα11 was removed and a KpnI restriction site (underlined) and, as a consequence, a two amino acid spacer (Gly-Asn) were introduced. Using the C-terminal primer 5′TTGTGCGGCCGCCGGTCACACCAGGTT-3, a NotI restriction site (underlined) was introduced downstream of the stop codon of Gα11. The amplified fragments digested with KpnI and NotI were subcloned into similarly digested pcDNA3 expression vector (Invitrogen). To obtain the various α1b-adrenoceptor-Gα11fusion proteins, the coding sequence of the wild-type or 3CAM (A293L, K290H, R288K), D142A, and A293E forms of the hamster α1b-adrenoceptor (all obtained from Susanna Cotecchia, Lausanne, Switzerland) were amplified by PCR. Using the amino-terminal primer 5′-GACGGTACCTCTAAAATGAATCCCGAT-3′, aKpnI restriction site (underlined) was introduced upstream of the initiator Met. Using the carboxyl-terminal primer 5′-GTCCCT GGTACC AAAGTGCCCGGGTG-3′, a secondKpnI restriction site (underlined) was introduced immediately upstream of the stop codon. Finally, the Gα11 constructs in pcDNA3 were digested withKpnI and ligated together with the PCR product of the α1b-adrenoceptor amplification, also digested with KpnI. The open reading frames thus produced represent the coding sequence of either α1b-adrenoceptor-Gα11, 3CAM α1b-adrenoceptor-Gα11, D142A α1b-adrenoceptor-Gα11, or A293E α1b-adrenoceptor-Gα11. Each was fully sequenced before its expression and analysis.
Transient Transfection of HEK293 Cells.
HEK293 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 0.292 g/l l-glutamine and 10% (v/v) newborn calf serum at 37°C in a 5% CO2 humidified atmosphere. Cells were grown to 60 to 80% confluency before transient transfection in 60-mm dishes. Transfection was performed using LipofectAMINE reagent (Invitrogen) according to the manufacturer's instructions.
[35S]GTPγS Binding.
[35S]GTPγS binding experiments were initiated by the addition of membranes containing 50 fmol of the fusion constructs to an assay buffer (20 mM HEPES, pH 7.4, 3 mM MgCl2, 100 mM NaCl, 1 μM guanosine 5′-diphosphate, 0.2 mM ascorbic acid, 50 nCi [35S]GTPγS) containing the indicated concentrations of receptor ligands. Nonspecific binding was determined in the same conditions but in the presence of 100 μM GTPγS. Reactions were incubated for 15 min at 30°C and were terminated by the addition of 0.5 ml of ice cold buffer, containing 20 mM HEPES (pH 7.4), 3 mM MgCl2, and 100 mM NaCl. The samples were centrifuged at 16,000g for 15 min at 4°C, and the resulting pellets were resuspended in solubilization buffer (100 mM Tris, 200 mM NaCl, 1 mM EDTA, 1.25% Nonidet P-40) plus 0.2% sodium dodecylsulfate. Samples were precleared with Pansorbin (Calbiochem, San Diego, CA), followed by immunoprecipitation with CQ antiserum (Mitchell et al., 1993). Finally, the immunocomplexes were washed twice with solubilization buffer, and bound [35S]GTPγS was measured by liquid-scintillation spectrometry.
[3H]Prazosin Binding Studies.
Binding assays were initiated by the addition of 3 μg of cell membranes to an assay buffer (50 mM Tris-HCl, 100 mM NaCl, 3 mM MgCl2, pH 7.4) containing [3H]prazosin (0.05–5 nM in saturation assays and 0.8 nM for competition assays) in the absence or presence of increasing concentrations of phenylephrine. Nonspecific binding was determined in the presence of 100 μM phentolamine. Reactions were incubated for 30 min at 30°C, and bound ligand was separated from free ligand by vacuum filtration through GF/B filters (Semat, St. Albans, Hertsfordshire, UK). The filters were washed twice with assay buffer, and bound ligand was estimated by liquid scintillation spectrometry.
Results
A fusion protein (α1b/Gα11) was generated in which the α subunit of the G protein G11 was linked in-frame with the C-terminal tail of the α1b-adrenoceptor from which the stop codon had been removed. The functionality of this protein has previously been established by monitoring the capacity of the agonist phenylephrine to elevate intracellular Ca2+levels in a fibroblast cell line in which the α subunits of both Gq and G11 had been eliminated by targeted gene disruption (Stevens et al., 2001). Immunoblotting membrane fractions of HEK293 cells positively transfected with this construct with an antiserum (CQ; Mitchell et al., 1993), which is directed against the C-terminal decapeptide common to Gαq and Gα11, identified the endogenously expressed forms of these G proteins. It also detected much higher levels of a doublet corresponding to the fusion protein (Fig. 1). Although the predicted molecular mass of the α1b-adrenoceptor and Gα11 is 56 and 43 kDa, respectively, the predominant form of the fusion protein migrated through SDS-PAGE with apparent molecular mass of 130 kDa. The doublet represents differentially glycosylated forms of the receptor-G protein fusion protein because pretreatment with N-glycosidase F compressed these to a single band, with an apparent molecular mass close to 100kDa (data not shown), as shown previously for a fusion protein between the human δ-opioid receptor and Go1α (Moon et al., 2001).
When membranes transiently expressing α1b/Gα11were subjected to a standard [35S]GTPγS binding assay, basal levels of [35S]GTPγS binding were high. Although this was stimulated significantly by the addition of a high concentration (10 μM) of the α1-adrenoceptor agonist phenylephrine (Fig. 2A), the relatively poor signal compared with basal is not suitable for detailed studies. Furthermore, the high basal levels of [35S]GTPγS binding were not a reflection of constitutive activity of the fusion protein because these were not different from those observed in membranes of mock transfected cells (Fig. 2A). However, when such samples were subsequently immunoprecipitated with antiserum CQ, the basal levels of [35S]GTPγS binding were very low in both the positively and mock-transfected cells. However, samples of the positively transfected cells that had been treated with phenylephrine now contained levels of [35S]GTPγS 30-fold higher than those that had not been exposed to the agonist (Fig. 2B).
Because HEK293 cells express both Gαq and Gα11, these results did not demonstrate conclusively that the agonist-induced binding of [35S]GTPγS was to the G protein element of the fusion protein. We thus constructed and expressed a version of the fusion protein in which the GPCR was linked to a form of the G protein (G208A Gα11) anticipated to be unable to release GDP and thus to bind [35S]GTPγS (Sprang, 1997). This construct bound the α1-adrenoceptor antagonist [3H]prazosin with the same high affinity as the fusion protein containing the wild-type G protein sequence (Table1) and was expressed almost as effectively (Table 1). After a membrane [35S]GTPγS binding assay and immunoprecipitation with antiserum CQ, basal levels of [35S]GTPγS binding were reduced compared with membranes expressing the fusion containing the wild-type G protein (Fig. 3), and the ability of phenylephrine to stimulate [35S]GTPγS binding was almost completely eliminated (Fig. 3). This was not a reflection that the α1b-G208A Gα11 fusion protein bound phenylephrine much more poorly than the α1b/Gα11 fusion protein. The capacity of phenylephrine to compete with [3H]prazosin for binding to the two fusion proteins was indistinguishable (Table 1). To monitor whether the receptor of the fusion protein was able to activate G proteins other than that linked to it, the α1b-G208A Gα11 fusion protein was coexpressed with excess wild-type Gα11. Now phenylephrine did indeed elevate binding of [35S]GTPγS in the immunoprecipitates from the α1b-G208A Gα11 fusion protein expressing membranes (Fig.3). This was to a substantially lower level than for the wild-type fusion protein and was dependent upon the presence of the α1b-adrenoceptor because immunoprecipitates from cells transfected to express Gα11 in the absence of the α1b-G208A Gα11 fusion protein contained no significant levels of [35S]GTPγS with or without treatment with phenylephrine (Fig. 3).
The [35S]GTPγS binding capacity of the α1b/Gα11 fusion protein was assessed in studies in which phenylephrine-stimulated [35S]GTPγS binding was measured in the presence of increasing concentrations of unlabeled GTPγS. When such data were corrected for dilution of specific activity and presented as a pseudo-saturation binding profile, the immunoprecipitate was able to bind 0.46 ± 0.01 mol of GTPγS/mol of fusion protein expressed in the membrane fraction (Fig. 4A). However, as parallel assessment of the immunoprecipitation efficiency of antiserum CQ indicated that only 45% of the expressed fusion protein was recovered with the amount of antiserum used (Fig. 4B), this corresponds to a true [35S]GTPγS binding capacity of the construct of 1.02 mol/mol. Such data confirmed that the full population of expressed construct was able to exchange and bind guanine nucleotides and was thus properly folded.
Because there was some evidence for a degree of constitutive activity of the α1b/Gα11 fusion protein to bind [35S]GTPγS (Figs. 2B and 3), we assessed whether this would be blunted by addition of the α1-adrenoceptor antagonist/inverse agonist phentolamine (Fig. 5). Although this was observed, the low levels of basal [35S]GTPγS binding made analysis difficult. The initially described constitutively active mutant of the α1b-adrenoceptor resulted from the replacement of a short segment of the third intracellular loop of this GPCR, with the equivalent section of the β2-adrenoceptor (Allen et al., 1991). We thus constructed a fusion protein between this form of the receptor, that we designate 3CAM (Stevens et al., 2000), and Gα11. This protein also bound [3H]prazosin with high affinity (Fig. 5B; Table1) but was expressed at significantly lower levels than the α1b/Gα11 fusion protein (Fig. 5B; Table 1). A similar feature has previously been observed for both the isolated 3CAM α1b-adrenoceptor (Lee et al., 1997) and a C-terminal GFP-tagged form of the 3CAM α1b-adrenoceptor (Stevens et al., 2000) compared with equivalent forms of the wild-type receptor. Now, however, addition of an equal amount of the 3CAM α1b-adrenoceptor-Gα11fusion protein to a [35S]GTPγS binding assay, followed by its immunoprecipitation, resulted in substantially higher levels of bound [35S]GTPγS than produced by the α1b/Gα11 fusion protein (Fig. 5C), demonstrating this form of the receptor to possess greater ability to stimulate its associated G protein in the absence of ligand than the wild-type form of the receptor. The presence of phentolamine (10 μM) during the [35S]GTPγS binding assay reduced incorporation of [35S]GTPγS into the 3CAM α1b/Gα11 fusion protein substantially (Fig. 5C), confirming that phentolamine acted as an inverse agonist for the 3CAM α1b-adrenoceptor. Phentolamine, however, was unable to reduce binding of [35S]GTPγS to the 3CAM α1b/Gα11 fusion protein to the very low levels of labeling of the wild-type α1b/Gα11 fusion protein (Fig. 5C). Such results indicate that phentolamine is a partial inverse agonist at the 3CAM α1b-adrenoceptor. Phentolamine bound the 3CAM α1b/Gα11 fusion protein with high affinity (Ki = 2.8 ± 0.1 × 10−8 M) (Fig. 5D), and concentration-response curves to phentolamine showed the inverse agonist effects to have an EC50 of 6.1 ± 0.7 × 10−8 M (Fig. 5E). Other ligands with affinity at the α1b-adrenoceptor, including WB4101, corynanthine, HV723, and 5-methylurapidil also functioned as inverse agonists at the 3CAM α1b/Gα11 fusion protein (Fig. 6). WB4101 and phentolamine were similarly effective with the others displaying lower extents of inverse activity.
Single point mutations close to the interface of transmembrane helix VI and the third intracellular loop of the α1b-adrenoceptor (Kjelsberg et al., 1992) and also at remote locations, such as the interface of transmembrane helix III and the second intracellular loop, are known to induce constitutive activity (Scheer et al., 1997). The best studied have been D142A and A293E. These mutations were introduced into the α1b/Gα11 fusion protein to allow direct comparisons of the level of constitutive activity and the effectiveness of phentolamine as an inverse agonist at the different mutants. Following transient expression of the wild-type α1b/Gα11 and each of 3CAM α1b/Gα11, D142A α1b/Gα11, and A293E α1b/Gα11, fusion protein saturation [3H]prazosin binding studies defined expression levels and allowed the same amounts of each fusion protein to be added to the assays. Each of the mutated fusion proteins bound substantially higher levels of [35S]GTPγS in the absence of agonist than did the wild-type (Fig. 7), with the 3CAM mutant binding significantly higher levels than the other mutants. Furthermore, phentolamine acted as a partial inverse agonist at each of the mutated fusion proteins (Fig. 7).
Many constitutively active mutant GPCRs remain responsive to agonist ligands, indicating that the mutations do not result in the adoption of a conformation equivalent to that produced by binding of the agonist. This was true for the 3CAM α1b/Gα11 fusion protein. Phenylephrine further stimulated the binding of [35S]GTPγS to this construct in a concentration-dependent manner (Fig. 8A), and as anticipated from previous studies on the isolated 3CAM α1b-adrenoceptor, the agonist displayed substantially greater potency at 3CAM α1b/Gα11 than at the α1b/Gα11 fusion protein (Fig. 8A). As with the 3CAM mutant, phenylephrine was also much more potent in stimulating the binding of [35S]GTPγS to the fusion proteins containing either the A293E or D142A single point mutants (Fig. 8A). Although each of the fusion proteins containing the constitutively active receptor mutants bound [35S]GTPγS in a ligand-independent manner, this was not to more than 20% of the level that was produced by addition of phenylephrine (Fig. 8B). Such results indicate that these mutants do not represent a very good model of the agonist-activated conformation of the receptor.
This difference in potency of phenylephrine at the fusions containing the mutant receptors reflected the differences in affinity of the agonist for these forms of the receptor (Fig. 8C). Although [3H]prazosin binding experiments on membranes expressing the various α1b/Gα11 fusion proteins indicated that antagonist binding affinity was not different from the wild-type (Table 1), phenyleprine displayed substantially higher affinity at the mutant constructs than at the wild-type (Fig.8C).
Discussion
Regulation of the binding of [35S]GTPγS is by far the most widely used assay to monitor ligand and receptor activation of heterotrimeric G proteins (Wieland and Jakobs, 1994). Despite this, there are many technical limitations that restrict further use of this approach. The most important is that while it can be extremely effective for the study of receptors that couple to the pertussis toxin-sensitive Gi subfamily G proteins, it has not historically provided good signal to noise for receptors that couple to members of the Gs or Gq family G proteins. At least in part, this reflects the relatively high rates of basal guanine nucleotide exchange of the Gi-G proteins compared with other family members. An inability to use this assay effectively is a particular limitation for receptors that couple to the Gqfamily G proteins. This reflects that assays for second messengers generated via this cascade (Ca2+, inositol phosphates, and diacylglycerol) cannot be easily performed on membrane preparations and thus for cells and tissues that have previously been harvested and stored.
Attempts to modify membrane [35S]GTPγS binding assays to produce reasonable signals for Gs or Gq-coupled receptors have adopted three strategies. The first of these has been to alter the temperature of the incubation or the concentration of guanine nucleotides and/or other assay reagents (Wieland and Jakobs, 1994). The second has been to express the receptor and G protein in systems, including insect Sf9 cells, that allow high levels of production of foreign proteins but that have low endogenous expression levels of Gi-like G proteins (Barr et al., 1997; Windh et al., 1999). The third, and most promising, incorporates a selective immunoprecipitation step at the end of the assay to eliminate the binding of [35S]GTPγS from other G proteins arising from basal exchange processes (DeLapp et al., 1999; Akam et al., 2001; Selkirk et al., 2001; Willets et al., 2001).
Although the immunoprecipitation approach has been used successfully, it retains a number of problems. The greatest of these is that little information is usually provided on the immunoprecipitation efficiency achieved, and thus it is difficult to compare results for receptor activation of different G proteins (DeLapp et al., 1999; Akam et al., 2001). Furthermore, because mutated receptors are regularly expressed at different levels than the wild-type, this poses difficulties in attempts to compare their G protein activation capabilities. In recent years, many groups have taken advantage of receptor-G protein fusions to overcome these limitations because the receptor and G protein are always present with the same 1:1 stoichiometry (Milligan, 2002a,b). By linking the same receptor to two different G proteins, it is then possible to make direct measures of the relative ability of ligands to activate each G protein. Indeed, using this approach, occupancy of the human δ-opioid receptor by the agonist [d-Ala2,d-Leu5]-enkephalin has been shown to result in 3 times greater activation of Gi1α than Go1α (Moon et al., 2001). By contrast, the maximal capacity of the μ-opioid receptor to activate Gi1α is the same as for the δ-opioid receptor (Moon et al., 2001). This approach is equally useful in examining the effects of receptor mutations. Although an Asp79Asn mutation in transmembrane helix II of the α2A-adrenoceptor reduces the maximal ability of this receptor to activate Gi1α by some 95%, this effect is overcome by addition of a reciprocal Asn422Asp mutation in transmembrane helix VII (Ward and Milligan, 2002).
Herein, we have combined the receptor-G protein fusion approach with an end of assay immunoprecipitation strategy to provide a highly reproducible means to monitor receptor-mediated regulation of the binding of [35S]GTPγS to Gα11. When we expressed an α1b/Gα11 fusion protein in HEK293 cells, prepared membranes, and then performed a [35S]GTPγS binding assay that was terminated by immunoprecipitation with an antiserum to the C-terminal decapeptide common between Gα11 and Gαq, very few counts were present in the immunoprecipitate. However, when the assay was performed in the presence of phenylephrine a 30- to 40-fold increase in [35S]GTPγS binding was achieved. However, as most cells endogenously express combinations of Gα11and Gαq, it was initially unclear if this agonist-induced binding of [35S]GTPγS was to the Gα11 of the fusion protein, endogenous Gα11/Gαq, or some combination thereof. However, when we expressed a fusion protein containing a G208A mutant of Gα11 that prevents GDP release, and thus [35S]GTPγS binding, phenylephrine had virtually no effect on the amount of [35S]GTPγS in the immunoprecipitate. It is not that the α1b/Gα11is unable to interact with endogenous G protein, however, as was shown in studies in which excess Gα11 was co-transfected along with the fusion protein (Fig. 3). These observations are consistent with previous work on both an α2A-adrenoceptor-Gi1α fusion protein (Burt et al., 1998) and following fusion of the neurokinin-1 receptor to both Gs and Gq (Holst et al., 2001). Thus, achieving a high ratio of fusion protein to endogenously expressed G protein is required to ensure that virtually all of the bound nucleotide is to the G protein linked to the receptor. This can be ensured by using an anti-receptor antibody rather than the anti-G protein antiserum for the immunoprecipitation. We have recently used this approach for fusion proteins between the α1b-adrenoceptor and forms of Gα11 mutated at the C-terminus such that they are no longer recognized by antiserum CQ (Liu et al., 2002). The only significant limitation has been that the immunoprecipitation efficiency of the anti-receptor antibody was lower than for the anti-G protein antiserum.
A key requirement for these studies was the ability to add equal amounts of different fusion proteins to the [35S]GTPγS binding assays. Prior specific [3H]prazosin binding assays on the membrane preparations allowed this, and the 1:1 stoichiometry of the receptor to G protein thus ensured that the same amount of G protein was present in each assay also. The observation that there was less [35S]GTPγS binding to the α1b/Gα11 fusion protein containing the G208A mutation in the absence of added ligand than to the α1b/Gα11 fusion protein suggested that the wild-type α1b-adrenoceptor displays a degree of constitutive capacity to activate Gα11. We thus tested if phentolamine would function as an inverse agonist and reduce basal levels of [35S]GTPγS binding to α1b/Gα11. As it did so, we constructed a series of further α1b/Gα11 fusion proteins that incorporated mutations in the receptor previously recognized to enhance constitutive activity. We reasoned that the [35S]GTPγS binding assay would thus provide a direct monitor of the degree of constitutive activity of these mutants that would neither be limited by potential saturation of signal due to amplification nor be affected by issues of receptor reserve.
The first studied constitutively active α1b-adrenoceptor was produced by substitution of a short segment of the third intracellular loop of this receptor with the equivalent section from the β2-adrenoceptor (Allen et al., 1991; Lee et al., 1997). Expression of a Gα11 fusion protein containing this form of the receptor followed by membrane preparation, [3H]prazosin binding, and addition of an equal amount to the [35S]GTPγS binding assay indeed demonstrated this construct to bind substantially higher levels of [35S]GTPγS in the absence of ligand than the fusion containing the wild-type receptor. Furthermore, addition of phentolamine produced a marked reduction in basal [35S]GTPγS binding to the 3CAM form of the fusion protein that corresponded to many more counts than were available for potential inhibition resulting from the weak constitutive activity of the fusion containing the wild-type receptor. Thus, fusion proteins containing constitutively active mutant GPCRs are more suited to the detection and analysis of ligands possessing inverse agonism.
A concern expressed about such fusion proteins is that they may not replicate the basic characteristics of coexpressed, but separate, receptors and G proteins. To address this in relation to the expected characteristics of the 3CAM α1b-adrenoceptor, we monitored both agonist and antagonist binding affinity and concentration-response curves for [35S]GTPγS binding. Both the affinity and potency of phenylephrine were some 50-fold higher for the fusion incorporating the 3CAM receptor. However, when assessed by the level of basal [35S]GTPγS binding, the degree of constitutive activity of both the 3CAM α1b-adrenoceptor and the other mutants was less than might have been anticipated from previous studies that measured inositol phosphate generation. For example, the capacity of agonist to further stimulate inositol phosphate production in COS-7 cells expressing the A293E α1b-adrenoceptor has often been noted to be rather small compared with the effect of the unliganded receptor (Scheer et al., 1996; Rossier et al., 1999). The implication that the mutated receptor provides a relatively good model of the agonist-occupied receptor has been important to the generation of computer models of the activated state of the receptor (Scheer et al., 1996). However, as noted earlier, second messenger and other downstream measures of constitutive activity are subject to amplification and possibly other elements of integrative regulation. This study thus represents the first direct analysis of the level of G protein activation produced by constitutively active mutants of a Gα11/Gαq-coupled receptor. This is a much more direct assessment of the conformational alterations related to G protein activation that are imbued by the mutations relative to those induced by agonist-occupancy of the receptor. However, these studies indicate that in the absence of agonist, the various mutants of the α1b-adrenoceptor used are actually rather poor at causing guanine nucleotide exchange on Gα11compared with the agonist and are at best able to produce some 20% of the effect (Fig. 8B). This might have been contentious if the G protein element of the fusion protein was not able to bind [35S]GTPγS quantitatively in response to agonist. However, by combining saturation [35S]GTPγS binding studies with measures of immunoprecipitation efficiency, we were able to demonstrate the capacity of these fusion proteins to bind the theoretical maximum of 1 mol of [35S]GTPγS/mol of G protein α subunit.
A number of blockers at the α1b-adrenoceptor were shown to be able to reduce the basal activation of [35S]GTPγS binding to Gα11, indicating them to be inverse agonists (Fig. 6). This characteristic has previously been noted for the isolated 3CAM receptor (Lee et al, 1997) and further validates the premise that such receptor-G protein fusions behave as for the isolated polypeptides. However, in all cases these ligands acted as partial inverse agonists because they were not able to reduce the signal to the basal signal produced by the wild-type receptor. Furthermore, the extent of inverse agonism produced by phentolamine was not substantially different between different CAM mutants of the α1b-adrenoceptor.
By developing a robust [35S]GTPγS binding assay appropriate for the Gq family G proteins and using a strategy in which modified forms of the receptor have access to the same number of G proteins, we have produced a range of novel observations on the level of constitutive activity of α1b-adrenoceptor mutants and the effectiveness of ligands as inverse agonists. This approach should be applicable to other receptors that couple selectively to Gqfamily G proteins and may provide a useful screen for inverse agonists at this class of receptors.
Acknowledgments
We thank Susanna Cotecchia (University of Lausanne, Switzerland) for cDNAs encoding the forms of the α1b-adrenoceptor and helpful discussions.
Footnotes
-
Financial support for this work was provided by the Medical Research Council.
-
Current address: Scottish Biomedical, Todd Campus, West of Scotland Science Park, Glasgow G20 OXA, Scotland, UK.
-
DOI: 10.1124/jpet.102.035501
- Abbreviations:
- GPCR
- G protein-coupled receptor
- GTPγS
- guanosine-5′-O-(3-thio)triphosphate
- CAM
- constitutively active mutant
- PCR
- polymerase chain reaction
- PAGE
- polyacrylamide gel electrophoresis
- WB4101
- 2-(2,6-dimethoxyphenoxyethyl)aminomethyl 1,4-benzodioxane
- HV723
- α-ethyl 3,4,5-trimethoxy-α-(3-[2-(2-methoxyphenoxy)ethyl]}propyl)benzeneacetonitrile fumarate
- Received February 28, 2002.
- Accepted April 9, 2002.
- The American Society for Pharmacology and Experimental Therapeutics