G Protein-Coupled Receptor Heteromers as New Targets for Drug Development

Abstract

We now have a significant amount of experimental evidence that indicatesthat G protein-coupled receptor (GPCR) oligomerization, including homo- andheteromerization, is a general phenomenon. Receptor heteromers possess unique biochemical characteristics that are demonstrably different from those of its individual units. These properties include allosteric modulation(s) between units, changes in ligand recognition, G protein-coupling and trafficking.The discovery of GPCR oligomers have been related to the parallel discovery such as bioluminescence, fluorescence and sequential RET (BRET, FRET and SRET, respectively), time-resolved FRET (T-FRET) and fluorescence recovery after photobleaching (FRAP) microscopy. However, RET techniques are difficult to implement in native tissues. For receptor heteromers, indirect approaches, such as the determination of a unique biochemical characteristic (“biochemical fingerprint”), permit their identification in native tissues and their use as targets for drug development. Dopamine and opioid receptor heteromers are the focus of intense research which is related to the possible multiple applications of their putative ligands in pathological conditions, which include basal ganglia disorders, schizophrenia and drug addiction.

I. Unique Biochemical Characteristics of GPCR Heteromers

G protein-coupled receptors (GPCRs) are classically believed to be functional monomeric entities. In fact, recent studies have shown that monomers of class A GPCRs (adrenergic β₂, rhodopsin, and opioid m receptors), when reconstituted in lipid vesicles, couple and activate their respective G proteins upon agonist binding.^1–3 Also monomeric rhodopsin in solution can activate its G protein transducin.⁴ Nevertheless, we now have a significant amount of experimental evidence that indicates that GPCR oligomerization, including homo- and heteromerization, is a general phenomenon (see below) and, in fact, it still needs to be determined if GPCR monomers are functionally present in the cellular plasma membrane.

There is some confusion about the terms used to describe GPCR oligomers. One of the main reasons has been the lack of knowledge about what constitutes a receptor unit. We should first consider the current definition of “receptor,” which is, “a signal transducing unit, a cellular macromolecule or an assembly of macromolecules that is concerned directly and specifically with chemical signaling between and within cells.”⁵ It is therefore redundant to talk about a “functional receptor.” Recommendations for definitions and criteria for identification of receptor heteromers have recently been proposed.⁵ First we should make a distinction between “heteromeric receptor” or “homomeric receptor” and “receptor heteromer” or “receptor homomer.” A heteromeric receptor is a “dimeric or oligomeric receptor for which the minimal functional unit is composed of two or more different subunits that are not functional on their own.” This definition applies to ligand-gated ion channels (ionotropic receptors) such as glutamate N-methyl-d-aspartate (NMDA) receptors and most nicotinic acetylcholine receptors.^6,7 The term also applies to some tyrosine kinase receptors, such as receptors for the glial cell line-derived neurotrophic factor (GDNF) family of ligands, in which one subunit is responsible for the association with the ligand and the other subunit for the catalytic response.⁸ Furthermore, there are also heteromeric GPCRs, such as the GABAB receptor, which is composed of at least two seven-transmembrane (7TM) protein units, GABAB₁ and GABAB₂, which constitute a (“functional”) receptor. Thus, the GABAB₁ subunit is responsible for ligand binding and the GABAB₂ subunit is responsible for G protein activation and signaling.⁹ In this case we have complete asymmetry in the functioning of both 7TM units in the heteromeric receptor. Some taste receptors, for which genetic deletion of one of the subunits leads to suppression of the receptor function,⁹ should also becalled heteromeric GPCRs. If the receptor subunits are identical, they constitute a “homomeric receptor.” This is the case for some ionotropic receptors, such as the α₇ nicotinic acetylcholine receptor.⁷

On the other hand, a receptor heteromer is “a macromolecular complex composed of at least two (“functional”) receptor units with biochemical properties that are demonstrably different fromthose of its individual components.”⁵ Thus, a receptor homomer is the same as receptor heteromer, but combining two or more identical (functional) receptor units. In a recent study, using a functional complementation assay, D₂ receptor homodimers with a single G protein were suggested to be a minimal signaling unit (homodimeric D₂ receptors), which is maximally activated by agonist binding to one of the protomer, whereas additional agonist or inverse agonist binding to the second promoter blunts or enhances signaling, respectively.¹⁰ This allosteric modulation of signaling results from a direct interaction of the receptor homodimer (or homodimeric receptor) with the G protein, rather than from a downstream effect.¹⁰ A similar situation of two (functional) protomers and one G protein is most probably found in a receptor heterodimer, but in this case, two different ligands interact in the heteromer, and allosteric interactions in the receptor heteromer have also been described.⁵ It is currently believed that an allosteric interaction in a receptor heteromer involves an intermolecular interaction in which binding of a ligand to one of the protomers changes the binding properties of another promoter.^5,11,12 The same is believed to occur in receptor homomers, which is translated into either a positive or a negative cooperativity in ligand–receptor binding.^13,14

In addition to allosteric modulations, GPCR oligomerization involves changes in ligand recognition, G protein-coupling and trafficking. Opioid receptor heteromers constitute an example of such changes in ligand recognition at the receptor heteromer. The opioid receptor subfamily comprises μ, δ and κ receptors, but at least two pharmacological “subtypes” with different affinities for different ligands have been identified for each of the three cloned receptors.¹⁵ The δ receptor can form heteromers with the μ^16,17 and also with the κ receptors.¹⁸ It seems that the two pharmacological subtypes of δ receptors, δ₁ and δ₂ receptors, correspond to a δ-μ receptor heteromer and a δ receptor homomer, respectively.¹⁹ With regard to changes in G protein coupling, taking into account the fact that recent models support the binding of only one G protein to two receptor units,^10,20 this means that a receptor heteromer will at least have to “decide” which G protein it should bind to as these receptors are usually coupled to different G proteins. Dopamine receptors are classified as D₁-like, with the D1 and D5 receptor subtypes, which usually couple to Gs/olf proteins; and D₂-like, with the D2 and D4 receptor subtypes, which couple to Gi/o proteins.²¹ Histamine H₃ receptors also couple to the Gi/o protein when not forming heteromers.²¹ Recent studies have shown that the D₁–D₃ receptor heteromer couples to the Gs/olf protein and the D₁–H₃ receptor heteromer couples to the Gi/o protein.^22,23 In some instances, the receptor heteromer can “choose” a completely new G protein. Thus, the D₁–D₂ and D₂–D₅ heteromers couple preferentially to Gq proteins.^24,25

There is experimental evidence that indicates that most GPCR oligomers form in the endoplasmic reticulum (ER) and that their synthesis is a ligand independent process. In fact, it has been shown that in some cases dimerization/oligomerization allows the 7TM protein or GPCR (depending on whether we are dealing with a potential subunit of a homomeric or heteromeric receptor or a unit of a receptor homomer or heteromer) to escape from its conditional retention in the ER. When expressed alone, the GABA_B1 subunit is retained in the ER because of the presence of an ER-retention sequence in the intracellular C-terminal tail. Coexpression of the GABA_B2 subunit results in an intermolecular interaction between both subunits that masks the retention motif and allows surface delivery of the heteromeric GABAB receptor.⁹ The same situation occurs with some adrenergic receptors, such as the adrenergic α_1D receptor, which needs to heteromerize with α_1B in order to be transported to the cell membrane.²⁶ ER-retention motifs can, in fact, be used to demonstrate heteromerization. For instance, the cell surface delivery of the chemokine CXC1 receptor was impaired when it was fused to an ER-retention motif of the adrenergic α_1C receptor, but ER retention was prevented by cotransfection with the close-related chemokine CXC2 receptor.²⁷ Once on the cell surface, many GPCRs internalize spontaneously or upon agonist binding. Indeed, the capacity of a selective ligand of one GPCR to cause cointernalization of a second coexpressed GPCR that does not bind the same ligand is used, experimentally, as evidence for heteromerization. For instance, in cells coexpressing chemokine CXC₄ and CC₅ receptors, their respective ligands (CXCL12 and CCL5, respectively) induced cointernalization of both receptors.²⁸

II. Energy Transfer-Based Techniques to Study GPCR Oligomerization

The use of the energy transfer-based techniques, bioluminescence resonance energy transfer (BRET) and fluorescence resonance energy transfer (FRET) has been fundamental in taking the issue of GPCR oligomerization to the forefront of GPCR research, providing evidence of the presence of an increasing number of GPCR receptor heteromers in living cells.^29–31 RET involves nonradioactive transfer of energy from a chromophore in an excited state, the “donor,” to a fluorescent “acceptor” molecule. In FRET both molecules are fluorescent, whereas in BRET the donor molecule is an enzyme which becomes bioluminescent upon catalyzation of its substrate. Importantly, the efficacy of energy transfer is inversely proportional to the sixth power of the distance between the donor and acceptor chromophores. This implies that range of energy transfer from the donor to the acceptor is below 10 nm.^29–31 When using this technology to study GPCR oligomerization, the donor and acceptor molecules are usually fused to the C-termini of GPCRs and cotransfected in mammalian cell lines. Since the diameter of the seven TM helical core is estimated to be about 5 nm, a positive energy transfer strongly suggests that oligomerization has occurred.^29–31

It is important to point out that BRET and FRET do not demonstrate the existence of physical contact between the fused proteins, but only show that there exists a very close proximity, which could of course depend on oligomerization. Nevertheless, when using BRET, physical contact can be practically shown by using several different experimental approaches. Those possibilities include BRET saturation assays, BRET competition assays and experiments that show ligand-promoted changes in BRET.³² In BRET saturation experiments, a constant amount of the donor fusion protein (usually a receptor fused to Renilla Luciferase (RLuc)) is coexpressed with increasing amounts of the acceptor fusion protein (usually a receptor fused to yellow fluorescent protein (YFP)). If there is oligomerization, saturation is reached when all receptor-Rluc molecules are specifically associated with their receptor-YFP counterparts. By contrast, if the BRET signal results from random collision promoted by high receptor density, a quasi-linear curve is obtained. In BRET competition assays, an untagged receptor A or B is coexpressed with the receptor A-Rluc and receptor B-YFP. Theoretically, if there is receptor A-receptor B oligomerization, then the BRET signal is specific and decreases as a consequence of the untagged receptor interacting with one of the fusion proteins and competing for the complementary BRET fusion protein. Finally, in some cases, when receptor A-Rluc and receptor BYFP are coexpressed, changes in BRET signal due to ligand binding to the heteromer can be detected, which can be indicative of ligand-induced conformational changes in the heteromer. This has been, for instance, observed in experiments with melatonin (MT) receptor heteromers.³³ FRET-based methods have also been introduced to monitor signal transfer within a receptor heteromer upon ligand binding.34 These techniques detect movements (conformational changes) within a GPCR homo- or heteromer by inserting fluorophores, for instance, cyan fluorescent protein (CFP) and YFP, in intracellular loops of both or just one of the protomers (intramolecular FRET and intermolecular FRET). A nice example is the recently reported application of this technique to the study of adrenergic α_2A-opioid μ receptor heteromer. In this case, when the double-inserted α_2A receptor was coexpressed with the μ receptor, the α_2A receptor decreases its response to noradrenaline in the presence of morphine.³⁴ This inhibitory effect was extremely fast (half-life of less than 500 ms), strongly suggesting that it was due to direct protein–protein interaction rather than competition for G-protein subunits.³⁴

Sequential-BRET-FRET (SRET) has been introduced to identify oligomers formed by three different proteins.³⁵ In SRET, the oxidation of an RLuc substrate by an RLuc-fusion protein triggers the excitation of the acceptor GFP² by BRET² and subsequent energy transfer to the acceptor YFP by FRET.³⁵ BRET² differs from BRET¹ in the acceptor, GFP² instead of YFP, and the substrate for Rluc which is DeepBlue C instead of coelenterazine. This technique has been used to demonstrate the heteromultimerization between adenosine A_2A, dopamine D₂ and cannabinoid CB₁ receptors³⁵ and between A_2A, D₂ and glutamate mGlu₅ receptors.³⁶ The presence of A_2A–CB₁–D₂ and A_2A–D₂–mGlu₅ heteromers was also demonstrated by a combination of bimolecular fluorescence complementation (BiFC) and BRET techniques.^36,37 BiFC is based on the principle that a fluorescent complex from nonfluorescent constituents can be produced if a protein–protein interaction occurs. In this technique, two receptors are fused at their C termini with either the N-terminal or the C-terminal fragments of YFP, respectively; and receptor heterodimerization leads to YFP reconstitution. Then, if there is heterotrimerization, BRET can be obtained when the cells also coexpress the third receptor, which is fused to Rluc.^36,37

The use of acceptor and donor molecules that are genetically fused to GPCRs in the classical FRETand BRET approaches can alter the functionality of the receptor heteromer under study. Furthermore, the fusion proteins can also be expressed in intracellular compartments, making it difficult to demonstrate that the resonance energy transfer resulted from a direct interaction of proteins at the cell surface. Finally, an additional limitation of the classical FRET is the low signal-to-noise ratio resulting from the intrinsic fluorescence of the cells and the overlap between the emission spectra of FRET donors and acceptors. Time-resolved FRET (TR-FRET) is a FRET variant designed to circumvent these problems.³⁸ TR-FRET is based on the prolonged fluorescence characteristics of certain rare earth tracers such as the lanthanide compounds, which enable the processes of excitation and detection to be separated temporally. Upon excitation, lanthanide compounds such as europium cryptate show a long-lived emission fluorescence, which can induce a long-lived emission from the acceptor molecule (e.g., Alexa Fluor 647, DY-647, or d2). As lanthanide compounds have very low fluorescence emission at the emission wavelength of the acceptor, the signal-to-noise ratio increases enormously. In TR-FRET, the acceptor and the donor fluorophore molecules are usually conjugated to antibodies against N-terminal-epitope-tagged sequences and membrane-impermeant antibodies are used to exclusively assess cell surface GPCR oligomerization. However, the bivalent nature of antibodies could potentially stabilize large complexes and the large size, and multiple labeling of the antibodies can easily increase FRET due to random collision. A combination of TR-FRET with snap-tag technology has been recently introduced.³⁹ A snap tag, which is two-thirds the size of YFP, is introduced at the N-terminal end of a GPCR. The snap tag is derived from O⁶-guanine alkyltransferase, which covalently reacts with benzyl guanine, and the nonpermeant benzyl guanine derivatives which are covalently labeled with a fluorophore compatible with TRFRET measurements are used. This technology has been used to demonstrate the existence of homomers of the heteromeric GABAB receptor and to confirm the predominant arrangement of glutamate mGlu receptors as homodimers.³⁹

III. Receptor Heteromers as Pharmacological Targets

Receptor heteromers, in particular, open up many opportunities in pharmacology, since they constitute new targets for drug development. An important step in this field has been the identification of receptor heteromers in native tissues. RET techniques can be applied only when studying receptor heteromerization in artificial cell systems. Direct identification of heteromerization can be achieved by taking advantage of selective probes (e.g., specific antibodies or labeled selective ligands) that can discriminate between the receptor heteromer and other configurations of the individual components. Bivalent ligands, with selective binding for each of the two protomers connected by a spacer of variable length, can become a good strategy, as was recently reported in the study of the A_2A–D₂ receptor heteromer.⁴¹ No specific receptor heteromer antibodies have yet been reported. An indirect but valid strategy is to determine a “biochemical fingerprint” of the receptor heteromer, that is a biochemical characteristic of a receptor heteromer, which can be used for its identification in native tissue.⁵ A strong suggestion that a biochemical fingerprint is specific for a receptor heteromer can be obtained by showing that it is abolished or altered when the heteromerization is disrupted, or alternatively when the quaternary structure of the heteromer is significantly modified without disrupting heteromerization. This, of course, could be shown with RET in artificial cell systems but requires identification of the domains or epitopes (of at least one of the receptors) that form the interaction surface in the heteromer. This is a new area of research which uses computation techniques and bioinformatics.⁴² These studies can guide experiments using mutated/chimeric receptors or enable design of peptides that can selectively occupy and disrupt the receptor heteromer interface.⁵

Opioid and dopamine receptor heteromers are presently the focus of intense research, which is related to the possible multiple applications of their putative ligands in pathological conditions, which include basal ganglia disorders, schizophrenia, drug addiction and pain. In some cases, the allosteric interactions in the receptor heteromer can be used as a strategy to modify the effect of already existing drugs. The best example is the addition of an A_2A receptor antagonist as a means to potentiate the antiparkinsonian effects of l-dopa.⁴³ A_2A–D₂ receptor heteromerization has been demonstrated in mammalian transfected cells with coimmunoprecipitation and FRET and BRET (reviewed in Ref. 44). By using computerized modeling, pull-down techniques and mass spectrometric analysis, it was shown that A2A–D₂ receptor heteromerizationdepends on an electrostatic interaction between an Arg-rich epitope located in the amino-terminal portion of the third intracellular portion of the D₂ receptor and a single phosphate group from a casein kinase phosphorylatable Ser localized in the distal portion of the carboxy-terminus of the A_2A receptor.⁴⁵ In vitro studies with peptides corresponding to both epitopes demonstrated that the Arg-phosphate interaction possesses a “covalent-like” stability. Hence, these bonds could withstand fragmentation by mass spectrometric collision-induced dissociation at energies similar to those that fragment covalent bonds.⁴⁶ The Arg-phosphate electrostatic interaction between epitopes located in intracellular domains is obviously not the only interaction responsible for A_2A–D₂ receptor heteromerization. Thus, a significant but not complete reduction of BRET was observed when transfected cells express mutated D₂ receptors that lack the key amino acids involved in the Arg-phosphate interaction,⁴⁵ indicating that other receptor domains are also involved. Most probably, transmembrane domains also play a role in A_2A–D₂ receptor heteromerization, as has been demonstrated for other GPCR homomers and heteromers.^47,48 Nevertheless, the significant modification of BRET with mutated receptors indicates that the Arg-phosphate interaction is necessary to provide the final quaternary structure of the heteromer, which in fact determines its function. Patch-clamp experiments in identified GABAergic enkephalinergic neurons demonstrated that a disruption of the Arg-phosphate interaction in A_2A–D₂ receptor heteromers (by intracellular addition of small peptides with the same sequence as the receptor epitopes involved in the Arg-phosphate interaction) completely eliminates the ability of the A_2A receptor to antagonistically modulate the D₂ receptor-mediated inhibition of neuronal excitability.⁴⁹

The above-mentioned antagonistic interaction between A_2A and D₂ receptors is most probably related to the existence of an allosteric modulation of the A_2A–D₂ receptor heteromer. In other cases, targeting both protomers in the heteromer can induce some trafficking-related unwanted effects. For instance, the opioid δ-μ receptor heteromer seems to be a better target than either the μ or the δ receptors alone, since blockade of the δ receptor decreases tolerance to the analgesic effects of the most used μ receptor agonist, morphine. In keeping with this, bivalent ligands (with a μ receptor agonist and a δ receptor antagonist attached by a spacer with variable length) are being developed.⁵⁰ It is important to note that the existence of the δ-μ receptor heteromer is currently the focus of a debate, since a recent study has questioned the colocalization of the δ and μ receptors,⁵¹ this being an obvious prerequisite for receptor heteromerization. Thus, using a recently generated δ receptor-eGFP knock-in mouse strain,⁵² δ ans μ receptor cellular colocalization in dorsal root ganglia was demonstrated using anti-GFP and anti-μ receptor antibodies; and was reported to be 5%.⁵¹ But given the increased level of receptor expression in these knock-in mice⁵² and the high avidity of the anti-GFP antibody as compared to the anti-μ receptor antibody, it is likely that the level of μ receptor coexpression with the δ eceptor was underestimated in this study. Furthermore, previous studies have found that the GFP tag at the C-terminus affects the maturation of the δ receptors⁵³ and that the levels of receptor attenuate the maturation of m receptors.⁵⁴ Finally, an important point to remember is that GPCR expression is altered during development and in pathology. Under these conditions, the level of receptor coexpression is likely to be significantly altered, making some particular GPCR heteromers attractive drug targets. The dopamine D₁–D₃ receptor heteromer is a putative example of a pathology-involved receptor heteromer. It is thus possible that an increase in the expression of D1–D3 receptor heteromers is directly involved in l-dopa-induced dyskinesia in patients with Parkinson’s disease.^22,55