WO2006035208A1

WO2006035208A1 - Method for measuring binding of a test compound to a g-protein coupled receptor

Info

Publication number: WO2006035208A1
Application number: PCT/GB2005/003688
Authority: WO
Inventors: Mark Samuel Jonathan Briggs; Albert Francis Santos
Original assignee: Ge Healthcare Uk Limited
Priority date: 2004-09-30
Filing date: 2005-09-27
Publication date: 2006-04-06
Also published as: EP1794587A1; US20070218520A1; GB0421693D0; US7674584B2

Abstract

The invention provides a method for measuring binding of a test compound to a G-Protein Coupled Receptor (GPCR). The invention also provides a method for identifying and measuring the effect that an agent has upon modulating the binding of a test compound to a G-Protein Coupled Receptor.

Description

Method for Measuring Binding of a Test Compound to a G-Protein Coupled

Receptor

Field of Invention

The invention relates to a method for measuring binding of a test compound to a G-Protein Coupled Receptor (GPCR). The invention also relates to a method for identifying and measuring the effect that an agent has on modulating the binding of a test compound to a GPCR.

Background to the Invention

G-Protein Coupled Receptors (GPCRs) modulate the response of many drugs, hormones and neurotransmitters in biology. Many disorders and diseases are equally focussed around GPCR function, with therapeutics based on altering the responsivity of GPCR function by use of small ligands or peptides, acting as either agonists or antagonists.

More than 30% of all currently prescribed pharmaceutical drugs involve

GPCR-mediated modulation, and more than 30% of all drug targets classes are aimed at understanding and modulating GPCR function.

GPCR structure and associated binding proteins Currently, there are approximately 400 known GPCRs, characterised historically by nomenclature into Type A, (rhodopsin like), Type B (calcitonin) and Type C (metabotropic). With the advent of the sequencing of the human genome, sequence analysis and homology searching implied the presence of at least 150- 200 GPCR-like proteins which currently possess no known endogenous ligands. These latter GPCRs are known as "orphan" receptors (OC3PCR', Howard et al. (2001 ) Trends in Pharm. Sd., 22, 132-140). In total, there are ~ 400-500 endogenous ligands known to function via characterised GPCRs, including those for recently "de-orphanised" GPCRs .

GPCRs are characterised by a conserved seven-transmembrane spanning motif, which comprise of protein helices linked by both intracellular and extracellular loop domains. The extracellular domains of GPCRs contain ligand docking (binding) sequences, and the intracellular loop domains (2nd and 3rd loop) are important docking sites for GPCR-associated proteins (Moro et al. (2003) Chem. Commun., 24, 2949-2956).

In Nature, the ligand-binding event, which occurs at extracellular binding sites on the GPCR, is transduced, postulated to be via resultant protein conformational shifts, into the intracellular matrix. The transduction mechanism is signalled via an "early event" intracellular exchange of guanosine diphosphate (GDP) for GTP (guanosine triphosphate). GDP is present at the "resting", or "ligand-unoccupied" state, and exchange for GTP occurs at the "active", or "ligand-occupied" state. The binding of either GDP or GTP occurs at defined sites of an intracellular heterotrimeric complex, known as the G-proteins, which comprise 3 subunits, Ga, Gβ and Gγ. GTP and GDP bind to the Ga subunit of the heterotrimer. The G-protein complex resides on the intracellular face of membranes and is closely associated with residues within the intracellular loop domain of the GPCR. Coupling of most G protein-coupled receptors to heterotrimeric G proteins involves the third intracellular loop and the proximal region of the carboxyl-terminal tail of the GPCR.

Upon binding of GTP to the Gq subunit, there is a resultant perturbation of the G-protein complex, which subsequently induces downstream transduction via effector systems (e.g. such as phosphatidylinositol 4,5-bisphosphate (PIP2) hydrolysis with subsequent changes in this instance, to intracellular Ca²⁺ (Ca²⁺O levels). Ligand activation may also involve intemalisation of the receptor (e.g. for downstream gene induction) or direct opening of ligand-gated ion channels. At a later stage, regardless of the precise mechanism of ligand activation, there is a need for the response to be attenuated . To "decouple" or "downregulate" the response of ligand binding to the GPCR as well as attenuate subsequent downstream transduction events, the GTP-bound to the Gq subunit is hydrolysed by endogeneous enzymes (GTPases) back to guanosine diphosphate (GDP), leading to functional reassociation of the G-proteins and dissociation of the ligand with a return to the "resting" or "ligand-unoccupied" state.

GPCR-active ligands, and GPCR action in general, can also be characterised by the nature of the transduction event linked to Ga functionality. Gq functionality has been shown to be linked to primary sequences. It therefore follows that G protein classes can also be defined according to the primary sequences of their Gσ subunits. This classification has lead to definition of 4 families: Gas, GoMo, Gaq and Gσi2/13.

• Gα_s (μ[cAMP] via adenylate cyclase activation)

Some GPCR-active ligands can be characterised by a downstream increase in intracellular concentration of adenylate cyclase, which leads to a subsequent increase in intracellular concentrations of the important second messenger, cyclic adenosine monophosphate (cAMP). Upon ligand binding, transduction in this case is via GDP<→GTP exchange at the Gα_s subunit.

• Gσj_/o (o[cAMP] via adenylate cyclase inhibition, or K⁺ channel modulation or phosphodiesterase (PDE) activation) - Upon ligand binding, transduction occurs via GDP<→GTP exchange at the Gαj subunit.

• Gσ_q (protein coupled, μCa²* via phospholipase C beta (PLC/?)) Transduction is via GDP«→GTP exchange at the Gα_q subunit. Response is pertussis toxin sensitive. PLC/? catalyses release of diacylglycerol (DAG) and inositol 1 ,4,5-phosphate (IP3) from inositol 4,5-diphosphate (PIP2). IP3 increase is linked to Ca²⁺ _J release. It has also been found that heterologous expression of Gσ-iβ (a member of Gα_q), can allow coupling of a wide range of GPCRs to PLC/? activity and allow measurements of Ca²⁺ _I flux.

• Gαi_2/13 (protein coupled, interacting with CI^' channels)

Transduction is via GDP<→GTP exchange at the G #_12/13 subunit

Methods of carrying out GPCR assays to measure ligand potency

There is a continuing desire within the pharmaceutical industry to exploit

GPCRs and orphan GPCRs as drug targets. Many methods have been used to measure GPCR activity and in vitro assays form an important part of high throughput screening strategies in the search for new GPCR-active ligands. Complementary technologies involve cell-based assay formats in which for example, Ca²⁺ flux measurement can be made within intact cells by use of calcium-sensitive fluorescent indicators. In the latter case, the use of sensitive detection platforms have been aided by the creation of chimeric G proteins (such as Gσ_s-Gα_q) or the heterologous expression of Gσi₆, to allow "forced coupling" of ligand response through PLCβ activation pathways, enabling a Ca²⁺ readout to be made (Milligan & Rees (1999) Trends in Pharm. Sci., 20, 118-124).

Traditional methods of carrying out GPCR assays involve use of radioactive ligands. These are employed in heterogeneous filter-based or homogeneous SPA-based (Scintillation Proximity Assay) assays. From these studies, the end user can obtain information on ligand potency by measurement of the radioactive counts on the filter (after separation of bound from free ligand) or directly on the SPA bead.

Use of radioactive [³⁵SIGTPYS To exploit the binding of GTP to Gq as a high sensitivity in vitro assay interrogation point, researchers have developed GTPase-resistant ("non- hydrolysable") analogues of GTP, with one of the most efficacious being radioactive [³⁵S]GTPyS (Milligan (2003) Trends in Pharm. Sd., 24, 87-90; Ferrer et al. (2003) Assay & DDT, 1, 261-273). When a non-radioactive ligand now binds to cell membranes carrying a functional GPCR [³⁵S]GTPyS is recruited to the G-protein Ga subunit. As [³⁵S]GTPyS is essentially "non-hydrolysable", the receptor/G-protein system is effectively "locked" in a ligand-occupied state. Now, radioactive filter counts or SPA counts of G-protein-bound [³⁵S]GTPyS allows the user to obtain information on both the ligand binding potency as well as the ligand efficacy. Use of [³⁵S]GTPyS in this manner means that, in essence, the user is carrying out an in vitro "functional assay". The GTP-probe is effectively acting as a post-binding event reporter, at an early position in the transduction process.

The need for homogeneous fluorescence assays for GPCRs Whilst inherently sensitive radioactive assays (heterogeneous and homogeneous format) have formed the bulk of generic in vitro screening assays for GPCRs, there has been a desire to move towards sensitive, non-radioactive, and in particular homogeneous assays (Kimble et al., (2003) Combin.Chem & High Thr. Screening, 6, 409-418). The latter assay formats are particularly amenable to miniaturisation and hence provide time and material cost savings. A robust signal which can be easily measured on a spectrophotometer, in particular an optical signal, would be of advantage. Fluorescence intensity measurements, and in particular Fluorescence Resonance Energy Transfer (FRET), would fulfil many desirable requirements for a suitable assay format.

FRET is a distance-related process in which the electronic excited states of two dye molecules interact without emission of a photon (Forster, T., "Intermolecular Energy Transfer and Fluorescence", Ann. Physik., Vol.2, p.55, (1948)). One result of this interaction is that excitation of a donor molecule enhances the fluorescence emission of an acceptor molecule. The fluorescence quantum yield of the donor is correspondingly diminished. For FRET to occur, suitably, the donor and acceptor dye molecules must be in close proximity (typically between 10-100 A), since energy transfer efficiency decreases inversely as the 6th power of the distance (r) between the donor and acceptor molecules.

In FRET, molecules which act as FRET "donors" are allowed to interact with molecules which act as FRET "acceptors". By donor, it is meant that the dye moiety is capable of absorbing energy from light and emits light at wavelength frequencies which are at least partly within the absorption spectrum of the acceptor. By acceptor, it is meant that the dye moiety is capable of absorbing energy at a wavelength emitted by a donor dye moiety.

If these donor and acceptors come into close contact within a critical distance, then FRET occurs and spectroscopic measurements taken at the emission wavelengths of the acceptor will give an indication of the magnitude of the FRET interaction. If the donor and acceptor fluors are allowed to come into close contact as a result of a biological interaction, then it follows that the magnitude of the FRET signal will be related to the magnitude of the biological interaction under scrutiny. Under suitable conditions, the closest molecular distances between the FRET partners can be calculated from the maximum FRET signal.

Fluorescent Analogues of GTP

There has always been a desire to develop non-radioactive (fluorescent) reporter analogues of [³⁵S]GTPyS. Many have been described in the literature, but most suffer from high rates of hydrolysis and/or poor affinity for the G-proteins (McEwen et al. (2001) Anal. Biochem., 291, 107-117); Korlach et al., (2004) Proc. Natl.Acad. ScL, 101, 2800-2805). There is, therefore, a need within the pharmaceutical industries for a hydrolytically stable fluorescent reporter analogue which has a high degree of affinity for the G-proteins. Such a reporter molecule is described herein and is the subject of the Applicant's (Amersham Biosciences UK Limited) co-pending patent application entitled 'Fluorescent Nucleotide Analogues' (priority claiming applications GB 0421691.7 and GB 0500504.6).

Prior art - examples of "intermolecular" GPCR FRET assays Both in vitro and cell-based GPCR FRET assays have been cited in the literature. The FRET interaction in these instances is between two interacting "partner" biological species (for example, proteins) with the "donor" and "acceptor" fluorescent molecules bound to their respective but separate, species. When the two biological partners interact, FRET can occur under controlled conditions.

As referred to herein, the term "intermolecular interactions" are described as those occurring between separate G-protein subunits, Gσ , Gβ and Gγ.

Leaney et al., (J. Biol. Chem. (2002) 277, 28803-28809) describes the potential use of cyan fluorescently tagged Gσ-protein subunits in FRET assays for investigating protein-protein interactions. Similarly, WO 03/008435 postulates on the use of Gσ-green fluorescent protein (GFP) constructs in FRET assays for screening for GPCR drug targets. A method for detecting ligand binding using a FRET assay based upon the interaction of a blue fluorescent protein -Gσ construct with a yellow fluorescent protein-Gσ construct is reported in WO 02/077200 for identifying proteins involved in olfaction.

Bunemann et al., (Proc. Natl.Acad. Sci. (2003) 26, 16077-16082) describe use of cloned fluorescent protein tagged G-proteins which were viably reconstituted into cultured host human embryonic kidney (HEK) cells. G-proteins were tagged with either cyan fluorescent protein (CFP) or yellow fluorescent protein (eYFP), namely, Gαi-eYFP, G/?1-CFP and/or Gγ2-CFP. FRET signals were observed that were ligand (agonist) dependent, and which were postulated to be as a result of G-protein conformational shifts in response to specific ligand binding, allowing a measure of both ligand binding potency as well as changes in intermolecular distances, as the ligand "on-off cycle progresses.

Potential limitations pertaining to this latter "intermolecular" strategy is the requirement for two (or more) species to i nteract at appropriate times, orientations and concentrations. Significant alteration of the endogenous G- proteins by attachment of large fluorescent proteins may well lead to perturbation of the binding events under investigation. Alternatively, random chemical labelling with smaller, low molecular weight (MW) fluorescent tags can be carried out, but this may also lead to perturbation of natural biological function due to for example, unwanted chemical modifications at key binding sites and subsequent attenuation of binding affinity. There is also a real possibility of an increase in non-specific binding interactions when more than one species is required for an interaction. Also, the creation of two or more binding partners each labelled with potentially large fluorescent proteins may for example, lead to severe steric interactions leading to an attenuated or anomalous FRET response.

In addition, two biological species (Ga and G/? or Gγ in the example cited) have to be labelled with FRET partners, and if the labelling is intrinsic, then suitable cloning vectors have to be constructed leading to the generation of two or more recombinant proteins. To counter this situation by use of extrinsic labelling strategies, each binding partner may have be individually and site- specifically chemically labelled, which may be cumbersome, due to the requirements of high chemo- and regio-selectivity control.

Blaesius et al., (Presentation Number 135 in Session on 'Assay Development and Validation Strategies', The Society for Biomolecular Screening - 7^th Annual Conference, 12 September 2001) and WO 2004/035614 describe another example of an "intermolecular" FRET assay for GPCRs (Figure 1), using engineered peptide affinity probes. The authors describe the use of novel biotinylated peptide affinity probes which differentiate the GTP-bound state from the GDP-bound state of Qa_\ or Go₅. By using a carboxy-terminal histidine-tagged (hisβ) reconstituted GPCR, they were able to show, upon addition of suitable GPCR ligands, a detectable FRET signal between streptavidin-europium (bound to the biotin affinity peptide) and an allophycocyanin (APC)-labelled anti-histidine antibody.

A limitation pertaining to this method is the fact that the peptide affinity probe is not covalently bound to the target Gσ subunit, and therefore the relative affinities of a set of peptide probes need to be carefully evaluated for each set of Gσ species. This is an important issue, as there are multiple types of Gas, Gσi/o and Gσq that have been identified. Screening for sets of peptide probes of sufficient affinity is a laborious process, involving many cycles of compound generation and screening, such as by phage display panning, as well as optimisation by evolved library strategies. Indeed, even after suitable screening, the affinity of the peptide probe may either be quite low or be too highly cross- reactive between Ga species, so precluding use in an assay.

In addition, both the nature and position of binding of the probe is unknown, which is not an ideal situation when trying to optimise probe design. Another disadvantage of having a non-covalent binding probe is the risk of facile perturbation of this probe-target interaction by other factors, such as putative drugs or buffer/detergent conditions.

"Intramolecular" GPCR FRET assays Work by Frang et ai, ('Homogeneous GTP binding assay for GPCRs based on TR-FRET', poster, SBS 9^th Annual Conference, Portland, Oregon), uses identical peptide affinity probe technology invented by Blaesius et al. (Karo Bio, USA Inc.) described above. Frang describes use of a biotinylated peptide sourced from Karo Bio, which is a peptide affinity probe that recognises the GTP- bound state of Go_\. FRET occurs as a result of interaction between streptavidin- europium donor label (bound to the biotin affinity probe) and a fluorescent GTP analogue (Alexa647-GTP), which acts as a FRET acceptor.

Although the FRET response is configured around a single biological molecular entity (Ga-, in this case), and can therefore be referred to as

"intramolecular", the arguments cited earlier against employing the non-covalent binding of a biotinylated peptide affinity probe are still pertinent. Indeed, the arguments can be applied to any non-covalent binding approaches using any other type of probe, such as an antibody or aptamer.

The present invention seeks to address the above problems which exist in the prior art and to provide methods for detecting binding of a test compound (or ligand) to a GPCR and methods of identifying agents which modulate the binding of test compounds (or ligands) to GPCRs.

Summary of the Invention

According to a first aspect of the present invention, there is provided a method for detecting binding of a test compound to a G-Protein Coupled Receptor (GPCR), the method involving a Ga subunit which comprϊ ses a covalently bound first tag capable of binding to a first detectable group and either a GTPase resistant GTP analogue having a second detectable group, or a GTPase resistant GTP analogue having a second tag capable of bi nding to a second detectable group.

In a second aspect of the present invention, there is provided a method for detecting binding of a test compound to a G-Protein Coupled Receptor (GPCR), the method involving a Ga subunit which comprises a covalently bound first detectable group and either a GTPase resistant GTP analogue having a second detectable group, or a GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group. Suitably, the method of the first or second aspect involves use of a reaction mixture comprising at least one of the following reagents i) a test compound ii) a first detectable group iii) a second detectable group.

Suitably, the method of the first aspect comprises the steps of i) contacting the GPCR with the test compound ii) contacting the Ga subunit with the GTPase resistant GTP analogue having a second detectable group iii) binding a first detectable group to the covalently bound first tag, and iv) detecting a signal.

wherein the sequence of steps ii) and iii) is interchangeable.

Suitably, the method of the second aspect comprises the steps of i) contacting the GPCR with the test compound ii) contacting the Ga subunit with the GTPase resistant GTP analogue having a second detectable group, and iii) detecting a signal.

Suitably, the method of the first aspect comprises the steps of i) contacting the GPCR with the test compound ii) contacting the Ga subunit with the GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group iii) binding a first detectable group to the first tag iv) binding a second detectable group to said second tag, and v) detecting a signal. wherein step i) is the first step and step v) is the last step, the sequence of steps ii)-iv) being irrelevant.

Suitably, the method of the second aspect comprises the steps of i) contacting the GPCR with the test compound ii) contacting the Ga subunit with the GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group iii) binding a second detectable group to said second tag iv) detecting a signal.

wherein the sequence of steps ii) and iii) is interchangeable.

Suitably, the signal is compared to a signal obtained in the absence of the test compound.

Suitably, the test compound is an organic or inorganic molecule. Preferably, the organic molecule is selected from the group consisting of peptide, polypeptide, nucleotide, polynucleotide, protein nucleic acid, saccharide, polyglyceride and small organic molecule.

Preferably, the test compound is a ligand.

By establishing the use of such a strategy within the context of the first and/or second aspect of the invention, the skilled person is, for example, able to screen a number of potential natural endogenous ligands for their ability to bind to, as well as modulate downstream transduction events at a specific subtype of GPCR polypeptides. An extension of this strategy is to carry out "de-orphanising" of orphan GPCRs, whereby an orphan GPCR can be assigned to its associated endogenous ligand from both a binding and functional aspect. Suitably, the ligand is known to bind to the GPCR polypeptide. Examples of such known ligands which bind to particular GPCR polypeptides are well known in the art.

Suitably, the method is a homogeneous method.

In a third aspect of the present invention, there is provided a method for detecting the effect an agent has upon modulating the binding of a test compound to a GPCR, said method comprising detecting the binding of said compound to said GPCR as hereinbefore described in the presence of the agent and comparing binding in the absence of the agent.

While binding of the test compound (or ligand) can be detected qualitatively, preferably binding is measured quantitatively.

By the use of such a strategy, the end user is able to assign a rank order of potency of a number of agents relative to the reference ligand. This enables identification of potentially suitable drug candidates which are able to modulate GPCR function. The agents may be single molecular entities or one of a member of a group or a cassette, for example from a natural product library or from a synthetic phage display library or from a chemically-synthesised library.

Suitably, the value of the test compound or ligand binding in the absence of the agent is already known. The term "value" can be taken to mean an actual measure of the binding affinity of a ligand, such as the molar affinity constant or molar dissociation constant. Preferably, the value of the binding in the absence of the agent is stored on an electronic database such as a computer. Optionally, the binding value in the absence of the agent may be stored on an optical database. Electronic and optical databases are configured such that data stored thereon are readily accessible, typically by data searching and 'mining' techniques well known in the art. Suitably, the first detectable group comprises a first binding moiety which specifically binds to the first tag.

Suitably, the second detectable group comprises a second binding moiety which specifically binds to the second tag.

Suitably, the covalently bound tag and the binding moiety are members of a specific binding pair, wherein each component has a specific binding affinity for the other. Preferably, the tag and the binding moiety are selected from the group consisting of biotin/steptavidin, biotin/avidin, biotin/neutravidin, biotin/captavidin, epitope/antibody, GST/glutathione, His-tag/Nickel, FLAG/M1 antibody, maltose binding protein/maltose, chitin binding protein/chitin, calmodulin binding protein/calmodulin (Terpe, 2003, Appl Microbiol Biotechnol, 60, 523-533), LumioTM reagents/Lumio™ recognition sequence. The Lumio™ reagents and recognition sequence (Cys-Cys-Pro-Gly-Cys-Cys) are available from Invitrogen Life Corporation, Carlsbad, CA, USA.

Preferably, the first or second tag is poly histidine, such as (His)β, and the first or second binding moiety is Nickel. More preferably, the first or second tag is FLAG and the first or second binding moiety is M1.

Most preferably, the first or second tag is biotin and the first or second binding moiety is selected from the group consisting of streptavidin, avidin, neutravidin and captavidin. Streptavidin has a high binding affinity (10^"14M - 10^"15M) for biotin which makes the biotin tag/steptavidin binding moiety particularly suited for the present invention. The preferred labelling position of the tag would be either at the C-terminus, or preferably at the N-terminus of the Ga subunit. One advantage of using biotin to label the Gσ subunit is the small size of biotin compared, for example, with a Green Fluorescent Protein (GFP). Using a small molecule such as biotin causes less perturbation to the biological system under evaluation, compared to use of larger tags such as fluorescent-labelled proteins. In addition, use of biotin enables detection with proteins such as avidin (AV) or streptavidin (SA), which have well-documented and very high affinities (10^"14M - 10^"15M) for biotin. Avidin or streptavidin will be suitably labelled with a fluorescent moiety, which can act as a FRET "donor".

Suitably, the first and second detectable group is selected from the group consisting of a fluorescent moiety, a phosphorescent moiety, a luminescent moiety, a bioluminescent moiety, a chemiluminescent moiety, an absorbent moiety, a photosensitizer and a spin label.

Suitably, the fluorescent moiety is detectable by its fluorescence properties selected from the group consisting of fluorescence emission intensity, fluorescence polarisation (FP), fluorescence lifetime (FL), fluorescence anisotropy (FA), and fluorescence resonance energy transfer (FRET).

Preferably, the first detectable group and the second detectable group comprise fluorescent moieties which form a FRET pair, wherein the FRET pair comprises a donor FRET label and an acceptor FRET label.

Suitably, the donor FRET label is a xanthine dye or a cyanine dye.

Alternatively, the donor FRET label is a luminescent d-block and f-block metal containing complex, co-ordination compound or organometallic species.

Suitably, the donor dye is a xanthene dye, rhodamine dye or a cyanine dye and the acceptor dye is a xanthene dye, rhodamine or a cyanine dye as described in pending US provisional patent application 60/564924 'Fluorescence Resonance Energy Transfer Enzyme substrates'.

In one embodiment, at least one of said donor and acceptor dye moiety is a cyanine dye. In another embodiment, the donor dye is a xanthene dye and the acceptor dye is a rhodamine dye. Suitable xanthene dyes include but are not limited to fluorescein and its derivatives and analogues, such as 5-carboxyfluorescein, 6-carboxyfluorescein and δ-carboxy-^.δ'-dichloro^'J'-dimethoxyfluorescein.

Suitable cyanine dyes include but are not limited to CyA (3-(ε- carboxypentyl)-3'-ethyl~5,5'-dimethyl oxacarbocyanine),^' Cy2 (3-(e- carboxypentyl)-3'-ethyl-oxa-carbocyanine), Cy3 (3-(ε-carboxypentyl)-1 '-ethyl- 3,3,3',3'-tetramethyl-5,5'-disulphonato-carbocyanine), Cy3.5 (3-(e-carboxypentyl)- I'-ethyl-S.S.S'.S'-tetramethyl^.δ^'.δ'^I .S-disulphonatoJdibenzo-carbocyanine), Cy5 (1-(ε-carboxypentyl)-r-ethyl-3,3,3',3'-tetramethyl-5,5'-disulphonato- dicarbocyanine), Cy5.5 (1-(e-carboxypentyl)-1 '-ethyl-3,3,3\3'-tetramethyl- 4,5,4',5'-(1 ,3-disulphonato)-dibenzo-dicarbocyanine), and Cy7 (1-(e- carboxypentylJ-i'-ethyl-S.S.S'.S'-tetramethyl-δ.δ'-disulphonato-tricarbocyanine).

The preferred label is Cy3B (i.e. 6,7,9, 10-Tetrahydro-2-carboxymethyl-14- sulphonato-16, 16,18, 18-tetramethyl-7aH-bisindolinium[3,2-a,3\ 2'-a]pyrano[3,2- c;5,6-c']dipyridin-5-ium trifluoroacetate) but options can include other suitable donor labels or fluors, such as europium or terbium containing species. To improve FRET efficiency, the preferred embodiment allows optimisation of the dye-labelling stoichiometry on steptavidin or avidin.

In a preferred embodiment, the first detectable group is Cy3B-streptavidin. A standard range of dye labelling is usually one to three dye molecules per molecule of streptavidin (or avidin).

Suitably, the acceptor dye is a rhodamine dye or a cyanine dye.

Suitable rhodamine acceptor dyes include but are not limited to: 5- carboxyrhodamine (Rhodamine 110-5), 6-carboxyrhodamine (Rhodamine 110-6), 5-carboxyrhodamine-6G (R6G-5 or REG-5), 6-carboxyrhodamine-6G (R6G-6 or REG-6), N.N.N'.N'-tetramethyl-δ-carboxyrhodamine, N,N,N\N'-tetramethyl-6- carboxyrhodamine (TAMRA or TMR), 5-carboxy-X-rhodamine and 6-carboxy-X- rhodamine (ROX). Other classes of dyes include BODIPY™, porphyrin dyes, rhodol dyes, oxazine dyes and perylene dyes.

Suitable cyanine dyes include but are not limited to CyA (3-(e- carboxypentyl)-3'-ethyl-5,5'-dimethyl oxacarbocyanine), Cy2 (3-(e- carboxypentyl)-3'-ethyl-oxa-carbocyanine), Cy3 (3-(e-carboxypentyl)-1 '-ethyl- 3,3,3' ,3'-tetramethyl-5,5'-disulphonato-carbocyanine), Cy3.5 (3-(e-carboxypentyl)- r-ethyl-3,3₎3',3'-tetramethyl-4,5,4',5'-(1,3-disulphonato)dibenzo-carbocyanine), Cy5 (1-(e-carboxypentyl)-1'-ethyl-3,3,3',3'-tetramethyl-5,5'-disulphonato- dicarbocyanine), Cy5.5 (1-(e~carboxypentyl)-1 '-ethyl-3,3,3',3'-tetramethyl- 4,5,4' ,5'-(1 ,3-disulphonato)-dibenzo-dicarbocyanine), and Cy7 (1-(e- carboxypentyO-i'-ethyl-S.S.S'.S'-tetramethyl-δ.δ'-disulphonato-tricarbocyanine).

Preferably, the cyanine dye is a pentamethine cyanine dye.

It will be understood by one skilled in the art, that specific FRET donor/acceptor pairs must be selected based upon their particular emission and absorption characteristics.

The preferred acceptor label is Cyδ.

Preferably, the GTPase resistant GTP analogue is Cyδ-GTP. Most preferably, the Cyδ-GTP analogue is a compound having formula I or II.

Formula I

Formula Il

In one embodiment, the GTPase resistant analogue comprises a biotin tag. In this embodiment, the second detectable group could comprise, for example, a Cy5-streptavidin dye wherein the streptavidin binding moiety specifically binds to the biotin tag.

In a preferred embodiment, the covalently bound first tag is biotin, the GTPase resistant GTP analogue is Cy5-GTP and the first detectable group is Cy3B-Streptavidin. Thus, upon addition of a GPCR-active ligand (which may be a known or Reference ligand or a suitable candidate) which binds to and activates the GPCR, Cy5-GTP is recruited to the Gσ subunit, whereupon the binding of Cyδ-GTP to the target subunit is maintained due to the GTPase- resistant properties of the specifically engineered GTP analogue. Subsequent addition of the Cy3B-Streptavidin detectable group can enable the "donor" fluor (Cy3B, in this case) to be in close proximity (for example within 5nm) of the bound "acceptor" fluor (Cy5-GTP, in this example). Upon light excitation at the wavelength of the donor fluor, FRET occurs which can be detected at the acceptor fluor emission wavelength.

In another embodiment, the covalently bound first detectable group is Cy3B and the GTPase resistant GTP analogue is Cy5-GTP.

The magnitude of the FRET response (signal in presence of ligand) minus "basal" (signal in absence of ligand) allows determination of the potency of binding and the efficacy of the GPCR ligand under investigation.

Suitably, the ligand is a natural substrate or a synthetic substrate. This

'known' ligand will have well-characterised and measured GPCR binding properties, such as binding affinity, on/off kinetic binding rates and pharmacological response.

Suitably, the agent, which may also be an unknown or candidate ligand in its own right, is selected from the group consisting of agonist, antagonist and inverse agonist. The agent may be applied in the form of a group, library or cassette containing a plurality of such agents. The agent could also refer to a suitable environmental stimulus, such as induction of changes in temperature, pressure, ionic strength and pH. In addition, the agent could comprise a chemical entity which does not operate via a GPCR transduction system, but which indirectly affects an aspect of GPCR functionality. An example of the latter may be a protein synthesis inhibitor, functional antibody, or gene "knockdown" reagent (such as sRNAi or an antisense gene). As described herein, an agonist is any ligand (especially a drug or hormone) that binds to a receptor to alter the proportion that is in an active form to elicit a biological response. An antagonist is described herein as any ligand that results in the inverse response to an agonist while an inhibitor is any agent that blocks the biological response generated by the agonist. An inverse agonist is a drug which acts at the same receptor as that of an agonist, yet produces an opposite effect. Inverse agonists are also referred to as negative antagonists.

Suitably, the agent is selected from the group consisting of organic molecule, inorganic molecule, ion and environmental stimulus. Preferably, the organic molecule is selected from the group consisting of peptide, polypeptide, nucleotide, polynucleotide, protein nucleic acid, saccharide, polyglyceride and small organic molecule.

Suitably, the method is conducted on a cellular membrane fraction. The method may also be conducted on living, intact cells.

The present invention offers advantages over those methods known in the art by employing site-specific labelling of G-protein subunits, which are generic to a range of G-protein subtypes, and which circumvents one of the key problems associated with studies involving singular reporting with labelled GTP analogues alone. These latter approaches can and do suffer from high "basal" (or "background") rates of exchange of GTP for GDP in the absence of a GPCR ligand (Milligan (2003) Trends in Pharrn. ScL, 24, 87-90). In addition, other membrane proteins, such as tubulin, can exchange GTP for GDP. Thus, in the absence of ligand, there will be a fixed degree of basal cellular labelling, such as on the Ga subunit of the G-proteins. What this can mean is that when a specific GPCR ligand is subsequently added, the ligand "effect" of GTP analogue recruitment to the Ga subunit can be "masked" by the high endogenous basal exchange, resulting in a poor signal to background. Basal binding is biased due to high expression levels of the Gj family of G-proteins in mammalian cell systems, as well as higher rate of basal GTP exchange at the Gα.j subunits relative to other Gα-protein families (G₅ and G_q/G₁₂,i₃). In many cell membranes and cellular systems employed in the art, there is always going to be a mixed population of G-protein families, reflecting combinations of relative expression levels. It thus becomes very difficult to measure specific ligand-based activation of the Gα_s and Gα_q/Gαi₂,i3 G-proteins, which constitute a very important aspect of GPCR activity and hence potential therapeutic intervention, because of the masking effect.

A key aspect of the present invention is that a secondary reporter moiety is directed onto a G-protein subunit of a specific family (e.g. Gα_q), via a site- specific covalent tag. This will allow FRET reporting activity only at that Gα_q - protein family to be recorded. This is because the positive FRET signal, after addition of a Gα_q -active ligand, will only be generated when the reporter fluor on the specific Gα_q -protein comes into close proximity with the secondary fluor on the GTP analogue (e.g. Figure 2). In the absence of ligand, although the other G-proteins (e.g. Gα_s and GOCJ) may have experienced high rates of basal GTP exchange (and hence be "labelled" by the GTP-fluor analogue), no FRET signal will be generated from these latter basal interactions. In yet another manifestation of FRET interactions, a choice of FRET partners may allow time-resolved analysis which may confer advantages depending on the type of study under investigation. Suitable dyes for time-resolved analysis include the acridone class of dyes, as described in WO 02/099424, and the qui nacridone class of dyes as described in WO 02/099432.

The prior art methods described above do not show site specificity, and are not generic in the sense that a non-covalently bound affinity probe or antibody (for example) would have to be sourced and screened for, in every case. In these latter cases, these alternative probes may well have varying affinities for their targets. When screening for samples with mixed target populations (as could be the case for Gσ).having probes with varying affinity may well "misreport" this population difference due to biasing of binding towards the higher strength binders.

The use of a single covalently bound tag (or probe) in the method of the present invention circumvents many of the problems associated with sourcing and testing non-covalently bound probes. Having a covalently bound probe removes the likelihood of perturbation of probe-target interactions, which is a risk when non-covalent probes are employed. These perturbations may arise from non-optimised assay conditions or from exogenous test compounds, leading to false positives or negatives.

Thus use of a generic and covalent tag allows for facile modifications of the reporter system, thus introducing an element of versatility into choice of reporters and detection systems as well as allowing optimisation of both the labelling content and the concentrations of the components.

The method of the present invention offers further advantages over the prior art methods in that it is both fluorescent and homogeneous in nature, requiring no separation steps, thus conferring convenience and sensitivity.

In a fourth aspect of the present invention there is provided a nucleic acid construct comprising a nucleic acid sequence encoding a Ga subunit polypeptide comprising a GTP binding site and the means to ligate, by chemical or enzymatic methods, a covalently bound tag, the tag being capable of binding to a first detectable group as described hereinbefore, or a covalently bound first detectable group as hereinbefore described.

According to a fifth aspect of the present invention, there is provided a vector comprising the nucleic acid construct as hereinbefore described. Suitably, the vector is a plasmid or a viral vector. Preferably, the viral vector is an adenoviral vector or a lentiviral vector. In a sixth aspect of the present invention, there is provided a cell transfected with a vector as hereinbefore described. If the vector is a plasmid vector, transfection can be achieved by methods which are well known in the art, such as DNA compaction, electroporation and the use of chemical carriers. If the vector is a viral vector, transfection is achieved by exploitation of the virus coat transfection recognition motifs targeting onto the host cell.

Figure 5 illustrates the process by which suitable host cells can be transfected with a viral vector according to the invention. In the example give, intein-mediated site-specific biotynlation of the Ga subunit is employed using the in vivo approach described by Lue et al. (J.Am.Chem.Soc. (2004) 126, 1055- 1062). Cells are transfected with a shuttle vector, comprising a DNA encoding a Gα-intein subunit, and a vector comprising viral DNA (see, for example, US6140087) to produce non-replicative viral particles containing DNA encoding for the Gα-intein subunit. The cells are further transfected with cDNAs encoding GPCR, Gβ and Gγ subunits. In vivo intein-mediated biotinylation of the Gσ subunit is achieved by the addition of a cysteine-containing biotin tag to the cells. The resulting cells comprise functional Ga, GB and GK subunits and GPCR polypeptide.

Suitably the cell is either a stable cell line or a transient cell line. The term 'stable cell line' is used to describe a cell line where the foreign DNA from the vector has been stably integrated into the cell genome and is replicated upon cell division, such that daughter cells also posses the foreign DNA. In contrast, a 'transient cell line' is a cell line in which the DNA has not been stably integrated into the genome, is only expressed in the transformed cell and is not replicated upon cell division.

Suitably, the cell further expresses a Gβ GPCR subunit and a Gγ GPCR subunit. Preferably, the cell is eukaryotic cell. Most preferably, the cell is a mammalian cell.

In a seventh aspect of the present invention, there is provided a Ga subunit polypeptide encoded by the nucleic acid as hereinbefore described or produced by the host cell as hereinbefore described.

In a eighth aspect of the present invention, there is provided a kit of parts comprising a Ga subunit polypeptide according to the seventh aspect of the invention and a GTPase resistant GTP analogue having a second detectable group, or a GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group.

Preferably, the kit further comprises a first detectable group which is Cy3B-streptavidin. More preferably, the second detectable group is Cy5.

In a ninth aspect of the present invention, there is provided a kit of parts comprising a vector according to the fifth aspect of the invention, and a GTPase resistant GTP analogue having a second detectable group, or a GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group.

Brief Description of the Drawings

Figure 1 illustrates a prior art 'intermolecular' FRET assay for GPCR ligand binding.

Figure 2 is a schematic representation of a FRET assay according to the present invention. Figure 3 is a plasmid vector map comprising a nucleic acid encoding a Ga subunit according to the present invention.

Figure 4a shows the nucleic acid encoding a Ga subunit according to the present invention.

Figure 4b shows the amino acid sequence of a Ga subunit according to the present invention.

Figure 5 is a schematic representation of a method of viral infection of a host cell and in vivo intein-mediated biotinylation of a Ga subunit according to the present invention.

Figure 6 displays typical results from a FRET assay according to the present invention.

Specific Description and Examples

The invention is illustrated by reference to the following examples.

1. Synthesis of Cv5 labelled GTP analogue: Cv5-C2-GTP (Compound 1 or Formula I above)

1.1 Synthesis of O⁵-f3-(2-aminoethylamino)-1 ,2,3-trihvdroxy-triphosphoryl- guanosine: Compound 2

The tetra-lithium salt of GTP (25mg, 0.048mmol) and N-(3- dimethylaminopropyl)-N'-ethylcarbodiinnicle hydrochloride (36mg, 0.190mmol) were dissolved in 4ml of 0.1 M triethanolamine hydrochloride buffer (pH 7.2) in a 25ml round-bottomed flask fitted with a magnetic stirrer bead. N-Fmoc-ethane- 1 ,2-diamine hydrobromide(113mg, 0.31 mmol) was suspended in 1.5ml of 1 ,4- dioxan and added to the stirred solution in the flask. DMF was added dropwise to the suspension in the flask until it became homogeneous. The solution was stirred at ambient temperature under an atmosphere of nitrogen for 20 hours. TLC (RP-18, 40:60 methanol :water) showed that all the GTP had reacted. The solution was evaporated to dryness under vacuum. 2.5ml of a mixture of 20:80 piperidine:DMF was added to the residue and the mixture stirred at ambient temperature for 15 minutes. The solution was then evaporated to dryness under vacuum. The residue was dissolved in water (10ml) and extracted with diethyl ether (2 x 10ml), the aqueous phase was then evaporated to dryness under vacuum. TLC (RP-18, 40:60 methanol:water) showed a single spot (R_f 0.75) which turned purple when sprayed with ninhydrin.

The residue was dissolved in water and purified by HPLC using a MonoQ™ 10/10 column (Amersham Biosciences) eluting with a gradient of water to 100% 0.5M triethylammonium acetate solution (pH 7.0) over 60minutes at a flow of 3ml/minute. Detection was at 260nm. The major product eluted after 36 minutes. This material was evaporated to dryness under vacuum and the residue dissolved in a minimal volume of water. This was further purified by reverse phase HPLC using a 250 x 10 mm Jupiter™ C-18 column (Phenomenex) eluting with 0.1 M triethylammonium acetate solution (pH 7.0) at a flow of

4ml/minute. Detection was at 260nm. A single peak eluted after 11.3 minutes. This material was evaporated to dryness under vacuum, the residue was dissolved in water and the process repeated several times to remove as much triethylammonium acetate as possible to give compound (2) as a colourless solid. Mass spectrometry (ES⁺) gave MH⁺ at 566 and MNa⁺ at 588. (C-₁₂H₂₂N₇CM₃P₃ requires 565). Calculated yield from absorption at 253nm was 7.1 mg (0.012mmol, 25%)

1.2 Synthesis of Cv5 labelled GTP analogue: Cv5-C2-GTP (Compound 1)

Compound 2 (2μmol) was dissolved in 0.2ml of water in a 1.5ml polypropylene V-vial. To this was added 100μl 0.1 M sodium bicarbonate solution followed by 200//I of a solution of 10mg (14.4μmol) Cy5-NHS ester in 1ml dry DMSO. The tube was placed on roller for 20 hours at ambient temperature. This material was purified by reverse phase HPLC using a 250 x 10 mm Jupiter C-18 column (Phenomenex) eluting with a gradient of 0.1 M triethylammonium acetate solution (pH 7.0) to 50% acetonitrile over 30minut.es at a flow of 4ml/minute. Detection was at 650nm. The component eluting after 20minutes was collected, then evaporated to dryness under vacuum to give compound 1 as a blue solid.

Mass spectrometry (ES⁺) gave MNa²⁺ at 614.7 and MNa⁺ at 1228.3 (C₄₅H₆₀N₉O₂OPaS₂ requires 1203)

Calculated yield from absorption at 649nm was 0.56mg (0.46μmol, 23%)

2. Synthesis of Cv5 labelled GTP analogue: Cv5-C6-GTP (Compound 3 or Formula Il above)

2.1 Synthesis of O⁵-f3-(6-aminohexylamino)-1 ,2,3-trihydroxy-triphosphoryl- quanosine: Compound 4

The tetra-lithium salt of GTP (25mg, 0.048mmol) was dissolved in 2ml of 0.1 M triethanolamine hydrochloride buffer (pH 7.2) in a 25ml round-bottomed flask fitted with a magnetic stirrer bead. To this was added N-Fmoc-hexane-1 ,6- diamine hydrobromide (42mg, O.IOmmol) dissolved in 1ml dry DMF. N-(3- dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (72mg, 0.37mmol) was dissolved in 1 ml of the triethanolamine buffer and added to the stirred solution in the flask. DMF was added dropwise to the suspension in the flask until it became homogeneous. The solution was stirred at ambient temperature under an atmosphere of nitrogen for 20 hours. TLC (RP-18, 40:60 methanol:water) showed that some of the GTP had still not reacted. A further portion of N-(3- dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (72mg, 0.37mmol) was added and stirring was continued for a further 20 hours. The solution was evaporated to dryness under vacuum. 5ml of a mixture of 20:80 piperidine:DMF was added to the residue and the mixture stirred at ambient temperature for 15 minutes. The solution was then evaporated to dryness under vacuum.

The residue was dissolved in water (10ml) and extracted with diethyl ether (2 x 10ml), the aqueous phase was then evaporated to dryness under vacuum. TLC (RP-18, 40:60 methanol :water) showed a single spot (R_f 0.75) which turned purple when sprayed with ninhydrin. The residue was dissolved in water and purified by HPLC using a MonoQ 10/10 column (Amersham Biosciences) eluting with a gradient of water to 100% 0.5M triethylammonium acetate solution (pH 7.0) over 60minutes at a flow of 3ml/minute. Detection was at 260nm. The major product eluted after 22 minutes. This material was evaporated to dryness under vacuum and the residue dissolved in a minimal volume of water. This was further purified by reverse phase HPLC using a 250 x 10 mm Jupiter C-18 column (Phenomenex) eluting with a gradient of 0.1 M triethylammonium acetate solution (pH 7.0) to 40% acetonitrile over 40 minutes at a flow of 4ml/minute. Detection was at 260nrn. A single peak eluted after 25 minutes. This material was evaporated to dryness under vacuum, the residue dissolved in water and the process repeated several times to remove as much triethylammonium acetate as possible to g ive compound 4 as a colourless solid.

Mass spectrometry (ES⁺) gave MH⁺ at 622.21. (C16H3₀N₇O13P3 requires

621.4) Calculated yield from absorption at 253nm was 3.5mg (0.007mmol, 15%)

2.2 Synthesis of Cv5 labelled GTP analogue: Cy5-C6-GTP (Compound 3)

Compound 4 (2.3μmol) was dissolved in 0.2ml of water in a 1.5ml polypropylene V-w^'al. To this was added 200μl 0.1 M sodium bicarbonate solution followed by 300μl of a solution of 10mg (14.4μmol) Cy5-NHS ester in 1ml dry DMSO. The tube was placed on roller for 20 hours at ambient temperature. The material was purified by ion exchange HPLC (Hiprep™ 16/10 DEAE FF column). The column was eluted with water from 0-9 minutes, then water to 35% 2M sodium chloride from 9-40 minutes at a flow rate of 5ml/minute. Detection was at 253 and 650nm. The major product eluted after 28minutes. This material was de-salted by reverse phase HPLC using a 250 x 10 mm Jupiter C-18 column (Phenomenex) eluting with O.1 M triethylammonium acetate solution (pH 7.0) at a flow of 4ml/minute. Detection was at 253 and 650nm. The major component was collected then evaporated to dryness under vacuum to give compound 3 as a blue solid.

Mass spectrometry (ES⁺) gave MH⁺ at 1260.3. (C₄₉H₆₇N9O2₀P3S₂ requires 1258.3)

Calculated yield from absorption at 649nm was 0.7mg (O.βμmol, 26%)

3. Synthesis of Streptavidin labelled Cv3B: Cv3B-streptavidin

Streptavidin (10 mg; Rockland Immunochemicals; SOOO-01; 16.3U/mg) was dissolved in NaHCO₃ buffer (2.5 ml; 0.1 M; pH 9.2).

The concentration of the solution was measured by UV spectroscopy

(E_28O ⁰"^{1 %}=3.17; concentration (mg/ml) = (OD_28O x dilution factor)/3.17) then adjusted by addition of NaHCO₃ buffer to form a concentration of 2 mg/ml.

Cy3B NHS ester (2 mg; Amersham Biosciences; PA63100) was dissolved in DMF (0.4 ml) then added to the streptavidin solution and stirred at room temperature for 1 h. The reaction mixture was transferred to a pre-soaked Slide- A-Lyzer^® dialysis cassette (3-15 ml; Pierce Biotechnology; 66425; 10K MW cutoff) and dialysed (0.15 M NaCI; 4 x 2 I then O.1 M PBS-azide; pH 7.2; 4 x 2 I) at 4 ⁰C for 48 h. The dye:protein ratio was determined by UV spectroscopy using the following equation:

Dye:Protein ratio = AWI 30000

A₂80-(A₅64 x 0.05)/20OOOO

to give 3.3 dyes/protein

[Cy3B Λ_max=564 nm; e₅₆₄=130000 M^"1crrf¹; correction factor = 0.05; streptavidin e₂₈₀=200000 M^"1 cm^"1; correction factor = 0.05]

The concentration of the solution was adjusted to 1 mg/ml with PBS azide buffer and BSA [100 mg, Rockland Immunochemicals; BSA-10] was added. After stirring for 5 min, 1 ml aliquots were dispensed into P87 vials and the product was isolated after freeze drying.

4. Preparation of Gg Subunits

Preparation of tagged Gσ subunits by biological means can be carried out by a variety of methods known to those familiar with the art of cDNA cloning, creating vector constructs from this cDNA, transfecting a suitable host cell with the latter construct and generating the required target protein. Many of the processes described above can be carried out by use of commercially available reagents and "kits". 4.1 Preparation of Isolated Ga Subunits

In a typical example, where Ga is to be tagged with biotin and the biotin- Gσ is required as an isolated product, the following procedure can be carried out:

The required cDNA for Ga ("gene of interest") can be obtained via a gene bank repository. This gene can be cloned using established techniques into ideally, a bacterial host (for example, a "midi-prep" procedure in E. coli host bacteria). A commercially available vector can be used for the required tag, and the use of restriction enzymes and ligases allows insertion of the gene of interest into the vector to create a vector-plasmid construct. This vector-plasmid construct can be further transformed into competent bacterial host cells (for example TOP 10 E. coli) for long-term storage on beads.

Figure 3 shows a plasmid vector according to the present invention used to transfect HEK 293 cells. The PinPoint™ Xa vector (Promega Corporation) was used to effect expression of biotin-tagged Gα_q subunϊt in E. coli. The DNA encoding the Gσ_q subunit was inserted downstream from the sequence encoding the Gα_q subunit that is biotinylated in vivo as the fusion protein is expressed. The recombinant plasmid was transformed in E. coli and protein production was induced by IPTG. Figure 4a & b show the nucleic acid and amino acid sequences of the Gα_q subunit. To generate the required tagged Gσ protein of interest (e.g. Gσ_q) a bead containing the adsorbed transformed bacterial host cells is allowed to grow using established methods (e.g. antibiotic selection procedures). Resistant clones are manually selected and allowed to expand into large scale (for example up to 1 litre) culture vessels in the presence of biotin. The culture broth is spun down and stored as a pellet. To isolate the desired tagged protein, an aliquot of the pellet is treated with a suitable lysis reagent, with or without ultra-sonication, followed by a two-stage purification procedure. The first stage is by use of commercially-available biotin affinity monomeric avidin resins and the second stage is by size exclusion chromatography. The desired end product (e.g. biotin-tagged Ga) can be assessed for purity by traditional means, such as by gel electrophoresis and for protein concentration, using commercially-available protein assay kits..

4.2 Preparation of Integral Ga Subunits

In a typical example where Gq is to be tagged with biotin, and the biotin- Ga is required as an integral part of an intact GPCR receptor system together with the associated G-proteins, the following procedure can be carried out:

The required cDNA for Ga "gene of interest" can be obtained via a gene bank repository. A gene construct (for example, Gα-intein, where the intein is preferably ligated onto the N-terminus of Ga), is cloned into a mammalian expression vector using standard commercially-available technology. Suitable mammalian host cells (for example, HEK 293 cells) are transiently transfected with the vector using commercially available chemical transfection technology. Alternatively, transfection may be carried out using viral delivery, where the viral particles would contain the intein-Gα cDNA contructs. Viral delivery allows potentially higher transfection efficiency rates.

Preferably, the host cells will contain cDNAs allowing constitutive expression of a desired GPCR system including associated G-proteins, ion- channels etc. In addition, there may be introduced methods of "repression", or "silencing" the endogenous Ga of the host cell genome so as to allow full expression of the exogenous tagged Ga protein. As an example, "null" mutants of Ga do exist which could be employed. After a suitable incubation period in selected growth media and under controlled conditions, tagging reagents are added (for example, a derivatised biotin-thioester) to the medium. Mammalian cells are then harvested and lysed under controlled conditions to generate membrane fragments which can be stored for later use in GPCR assays. The latter fragments will contain membrane-associated GPCR and associated G- proteins, where Ga carries a suitable tag, and which should allow fully reconstituted GPCR activity to be monitored in the presence and absence of modulating agents, such as drugs, ions and environmental changes.

5. Assay Methods for Measuring Ligand Binding to GPCR

5.1 Assays involving Isolated Ga Subunits

In a typical example where Gq is to be tagged with biotin, and the biotin- Gσ is required as an isolated product for optimisation experiments (e.g. to optimise dye concentrations in a FRET assay format), the following procedure can be carried out:

The assay is set up in black 96-well microtitreplates, usually to obtain data sets of triplicate values, in a final reaction volume of 100μl. The concentrations of the 3 key components are as follows:

Biotin-Gσ subunit, purified (MW ~ 40,000 to 48,000, typically 4pmol/well), "donor" Cy3B-streptavidin (typically 2OnM final) (2pmol/well) and "acceptor" Cy5- GTP (typically 4OnM; 4pmol/well final). The binding buffer employed is typically 2OmM HEPES, 3mM MgCI₂, 10OmM NaCI + 100 μg/ml saponin, pH 7.5. Other buffers can include TRIS, with varying concentrations of MgCI₂, NaCI and detergents.

To the assay plate wells is added assay buffer (60μl-*100μl), followed by biotin-Gσ 10μl) (4pmol/well) and, where appropriate, GTPγS; (10μl) (typically 100 μM; 10μM final) (for non-specific binding "NSB" wells), Cy3B-SA (10μl) (20OnM, 2OnM; 2pmol final), Cy5-GTP (10μl) (40OnM, 4OnM; 4pmol final). The plate is gently shaken on a plate shaker for 30-45minut.es at room temperature.

Detection is carried out in a fluorescent Plate detector, with filter settings at 531 nm (excitation, bandwidth 25nm) and 665nm (emission, bandwidth 7.5nm). Figure 6 shows the result of a typical experiment carried out with 2 pmol/well of Cy3B and 2pmol/well of Ga subunit and a concentration range of Cy5-GTP (1-4 pmol/well). As can be seen, increasing the concentration of the Cy5-GTP acceptor results in an increasing fluorescent signal.

In this assay there is no associated membrane present nor accompanying integral GPCR. Therefore ligand binding events or perturbation of those events by agents, for example, cannot be carried out in this isolated and "decoupled" system. However, the assay can be used to test (for example) the integrity of the FRET system or to optimise concentrations of the FRET dyes, and so presents a useful approach to rational assay design.

5.2 Assays using Integral Ga Subunits

In a typical example, where Ga is to be tagged with biotin and the biotin- Ga is required as an integral part of an intact GPCR receptor system together with the membrane-associated G-proteins, the following assay can be carried out using, for example, cell-membrane extracts:

The assay is set up in black 96-well microtitreplates, usually to obtain data sets of triplicate values, in a final reaction volume of 100μl. A reference ligand (10μl) is plated out into appropriate wells at a suitable final concentration range of typically 1-10OnM. In the same wells is added either zero agent (10μl buffer only) or an increasing concentration of agent (10μl; typically a range of 1nM-10μM). Chosen microtitreplate wells will also contain excess non-labelled GTPγS (10μl) (typically 100μM; 10μM final) to assess the degree of non-specific binding (NSB). Membrane containing fully reconstituted GPCR of interest with associated G- proteins (typically 1-4pmol receptor/well), part of which will consist of biotin- tagged Ga subunit (10μl) is added. This is followed by "donor" Cy3B-streptavidin (10μl) (typically 2OnM final; 2pmol/well) and "acceptor" Cy5-GTP (10μl) (typically 4OnM final; 4pmol/well). Volume of the wells is made up to 100μl with assay binding as appropriate. The binding buffer typically employed is 2OmM HEPES, 3mM MgCI₂, 10OmM NaCI + 100μg/ml saponin, pH 7.5. Other buffers can include TRIS, with varying concentrations of MgCI₂, NaCI and detergents.

The plate is gently shaken on a plate shaker for 45minut.es at room temperature. These conditions can vary. Detection is carried out in a fluorescent plate detector, with filter settings at 531 nm (excitation, bandwidth 25nm) and 665nm (emission, bandwidth 7.5nm).

This assay, with fully integrated and reconstituted GPCR containing biotin- tagged Gσ, allows rank order of potency profiling of a range or agents when measured against a reference ligand or group of reference ligands.

It will be understood that the skilled person can use this assay to generate panels of any reconstituted GPCR of interest with combinations of Gσ, GB and Gγ subunits, where the Gσ is specifically tagged as herein before described.

If this GPCR of interest is a GPCR whose associated endogenous ligands are unknown (the so-called "orphan" receptors), then by use of appropriately- sourced cDNA to generate reconstituted GPCR and associated G-proteins with incorporated tagged-Gσ, the assay method allows the "screening" of that GPCR and assignment of function to a particular endogenous ligand chosen from a "pool" of biological extracts from a relevant source. This process of assignment of endogenous ligand to a GPCR which has previously unknown associated endogenous ligands, is known as "de-orphanisation". The assay method further allows assignments of endogenous ligands with the most noted biological effects to a particular GPCR. This information on both the target GPCR as well as its endogenous ligands, allows a more rational approach to improved understanding of the pharmacology of the GPCR and associated biological pathways, a clearer understanding of the role of the GPCR/endogenous ligand in a disease process, and a more rational approach to drug design.

Claims

1. A method for detecting binding of a test compound to a G-Protein Coupled Receptor (GPCR),

characterised in that said method involves a Gσ subunit which comprises a covalently bound first tag capable of binding to a first detectable group,

and

a GTPase resistant GTP analogue having a second detectable group,

or

a GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group.

2. A method for detecting binding of a test compound to a G-Protein Coupled Receptor (GPCR),

characterised in that said method involves a Ga subunit which comprises a covalently bound first detectable group,

and

a GTPase resistant GTP analogue having a second detectable group,

or

3. The method according to either of claims 1 or 2, wherein said method involves use of a reaction mixture comprising at least one of the following reagents i) a test compound ii) a first detectable group iii) a second detectable group.

4. The method of either of claims 1 or 3, said method comprising the steps of i) contacting the GPCR with the test compound ii) contacting the Gσ subunit with said GTPase resistant GTP analogue having a second detectable group iii) binding a first detectable group to the covalently bound first tag iv) detecting a signal.

wherein the sequence of steps ii) and iii) is interchangeable.

5. The method according to either of claims 2 or 3, said method comprising the steps of i) contacting the GPCR with the test compound ii) contacting the Gσ subunit with the GTPase resistant GTP analogue having a second detectable group iii) detecting a signal.

6. The method according to either of claims 1 or 3, said method comprising the steps of i) contacting the GPCR with the test compound ii) contacting the Gσ subunit with the GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group iii) binding a first detectable group to the first tag iv) binding a second detectable group to said second tag v) detecting a signal. wherein step i) is the first step and step v) is the last step.

7. The method according to either of claims 2 or 3, said method comprising the steps of i) contacting the GPCR with the test compound ii) contacting the Ga subunit with the GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group iii) binding a second detectable group to said second tag iv) detecting a signal.

wherein the sequence of steps ii) and iii) is interchangeable.

8. The method according to any preceding claim, wherein the signal is compared to a signal obtained in the absence of the test compound.

9. The method according to any preceding claim, wherein the test compound is a ligand.

10. The method according to any preceding claim, wherein the method is a homogeneous method.

11. A method for detecting the effect an agent has upon modulating the binding of a test compound to a GPCR, said method comprising detecting the binding of said compound to said GPCR according to the method of any preceding claim in the presence of the agent and comparing binding in the absence of the agent.

12. The method according to any preceding claim, wherein binding is measured quantitatively.

13. The method according to claim 11 , wherein the value of said binding in the absence of the agent is already known.

14. The method of claim 13, wherein said value of binding in the absence of the agent is stored on an electronic or optical database.

15. A method according to any preceding claim, wherein said first detectable group comprises a first binding moiety which specifically binds to the first tag.

16. A method according to any preceding claim, wherein said second detectable group comprises a second binding moiety which specifically binds to the second tag.

17. A method according to any preceding claim, wherein said first tag and/or second tag and said first and/or second binding moieties are members of a specific binding pair.

18. A method according to claim 17, wherein said specific binding pair is selected from the group consisting of biotin/steptavidin, biotin/avidin, biotin/neutravidin, biotin/captavidin, epitope/antibody, GST/glutathione, His tag/Nickel, FLAG/M1, maltose binding protein/maltose, chitin binding protein/chitin, calmodulin binding protein/calmodulin, and LumioTM reagents/LumioTM recognition sequence.

19. A method according to either or claims 17 or 18, wherein said first or second tag is biotin and said first or second binding moiety is selected from the group consisting of streptavidin, avidin, neutravidin and captavidin.

20. A method according to either of claims 17 or 18, wherein the first or second tag is poly histidine and the first or second binding moiety is Nickel.

21. A method according to either of claims 17 or 18, wherein the first or second tag is FLAG and the first or second binding moiety is M1.

22. A method according to any preceding claim, wherein the first and second detectable group is selected from the group consisting of a fluorescent moiety, a phosphorescent moiety, a luminescent moiety, a bioluminescent moiety, a chemiluminescent moiety, an absorbent moiety, a photosensitizer and a spin label.

23. The method according to claim 22, wherein said fluorescent moiety is detectable by its fluorescence properties selected from the group consisting of fluorescence emission intensity, fluorescence polarisation (FP), fluorescence lifetime (FL), fl uorescence anisotropy (FA), and fluorescence resonance energy transfer (FRET).

24. The method according to claim 23, wherein the first detectable group and the second detectable group comprise fluorescent moieties which form a FRET pair.

25. A method according to claim 24, wherein said FRET pair comprises a donor FRET label and an acceptor FRET label.

26. A method according to claim 25, wherein said donor FRET label is a xanthine dye or a cyanine dye.

27. A method according to claim 26, wherein said cyanine dye is a Cy3B dye.

28. A method according to any preceding claim, wherein the first detectable group is Cy3B-streptavidin.

29. A method according to any of claims 24 to 28, wherein the acceptor dye is a rhodamine dye or a cyanine dye.

30. A method according to claim 29, wherein the cyanine dye is a pentamethine cyanine dye.

31. A method according to claim 30, wherein the cyanine dye is Cy5.

32. A method according to any preceding claim, wherein the GTP'ase resistant GTP analogue is Cy5-GTP.

33. A method according to any preceding claim, wherein the GTP'ase resistant GTP analogue is a compound having a formula selected from:

and

34. A method according to any of claims 1-3, 6-19 and 22-31 , wherein the GTP'ase resistant GTP analogue comprises a biotin tag.

35. A method according to any of claims 1-3, 6-19 and 22-31 , wherein the second detectable group is a Cy5-streptavidin dye.

36. A method according to any of claims 1 , 3, 4, 6-19 and 22-33, wherein the covalently bound first tag is biotin, the GTP'ase resistant GTP analogue is Cy5-

GTP and the first detectable group is Cy3B-Streptavidin.

37. A method according to any of claims 2, 3, 5, 8-19, 22-27 and 29-33, wherein the covalently bound first detectable group is Cy3B and the GTP'ase resistant GTP analogue is Cy5-GTP.

38. A method according to any of claims 11 to 37, wherein said agent is selected from the group consisting of agonist, antagonist and inverse agonist.

39. A method according to any of claims 11 to 38, wherein the agent is selected from the group consisting of organic molecule, inorganic molecule, ion and environmental stimulus.

40. A method according to claim 39, wherein said organic molecule is selected from the group consisting of peptide, polypeptide, nucleotide, polynucleotide, protein nucleic acid, saccharide, polyglyceride and small organic molecule.

41. A method according to any preceding claim wherein said method is conducted on living, intact cells.

42. A method according to any of preceding claim, wherein the method is conducted on a cellular membrane fraction.

43. A nucleic acid construct comprising a nucleic acid sequence encoding a Gσ subunit polypeptide comprising a GTP binding site and the means to ligate by chemical or enzymatic methods

a covalently bound tag, said tag being capable of binding to a first detectable group according to any preceding claim,

or

a covalently bound first detectable group according to any preceding claim.

44. A vector comprising the nucleic acid construct of claim 43.

45. A vector according to claim 44, wherein said vector is a plasmid or a viral vector.

46. A vector according to claim 45, wherein said viral vector is an adenoviral vector or a lentiviral vector.

47. A host cell transfected with a vector according to any of claims 44 to 46.

48. A host cell according to claim 47, further expressing a GPCR together with associated G-protein subunits and integral ion channels.

49. A host cell according to either of claims 47 or 48, wherein said cell is an eukaryotic cell.

50. A host cell according to claim 49, wherein said cell is a mammalian cell.

51. A host cell according to any of claims 47 to 50, wherein said cell is a stable cell line or a transient cell line.

52. A Gσ subunit polypeptide encoded by the nucleic acid of claim 43 or produced by the host cell of any of claims 47 to 51.

53. A kit of parts comprising a Gσ subunit polypeptide according to claim 52 and a GTPase resistant GTP analogue having a second detectable group, or

54. A kit of parts according to claim 53 further comprising a first detectable group which is Cy3B-streptavidin.

55. A kit of parts according to claim 53 or 54 wherein the second detectable group is Cy5.

56. A kit of parts comprising a vector according to any of claims 44 to 46 and

a GTPase resistant GTP analogue having a second detectable group, or a GTPase resistant GTP analogue having a second tag capable of binding to a second detectable group.

57. A kit of parts according to claim 56 further comprising a first detectable group which is Cy3B-streptavidin.

58. A kit of parts according to claim 56 or 57 wherein the second detectable group is Cy5.