MINI-REVIEW

ARE G PROTEIN-COUPLED RECEPTOR HETERODIMERS

OF PHYSIOLOGICAL RELEVANCE?—FOCUS

ON MELATONIN RECEPTORS

Angelique Levoye,1 Ralf Jockers,1 Mohammed A. Ayoub,1

Philippe Delagrange,2 Egemen Savaskan,3 and Jean-Luc Guillaume1

1Department of Cell Biology, Institut Cochin, Universite Paris-Descartes, Faculte deMedecine, Paris, France; Departement de Pharmacologie Moleculaire, Institut deGenomique Fonctionnelle, Universites Montpellier 1 et 2, Marseille, France2Institut de Recherches SERVIER, Suresnes, France3Psychiatric University Clinic, Basel, Switzerland

In mammals, the circadian hormone melatonin targets two seven-transmembrane–spanning receptors, MT1 and MT2, of the G protein-coupled receptor (GPCR)super-family. Evidence accumulated over the last 15 yrs convincingly demonstratesthat GPCRs, classically considered to function as monomers, are actually organizedas homodimers and heterodimerize with other GPCR family members. Thesedimers are formed early in the biosynthetic pathway and remain stable throughoutthe entire life cycle. A growing number of observations demonstrate that GPCR oligo-merization may occur in native tissues and may have important consequences onreceptor function. The formation of MT1 and MT2 homodimers and MT1/MT2

heterodimers has been shown in heterologous expression systems at physiologicalexpression levels. Formation of MT1/MT2 heterodimers remains to be shown innative tissues but is suggested by the documented co-expression of MT1 and MT2 inmany melatonin-sensitive tissues, such as the hypothalamic suprachiasmatic nuclei,retina, arteries, and adipose tissue. Considering that multiple GPCRs are expressedsimultaneously in most cells, the possible engagement into heterodimeric complexeshas to be considered and taken into account for the interpretation of experimentaldata obtained from native tissues and knockout animals.

Keywords Melatonin, Circadian rhythm, G protein-coupled receptors

ESTABLISHMENT OF GPCR DIMERIZATION

The neurohormone melatonin is an important regulator of seasonalreproduction and circadian rhythms. Melatonin is synthesized mainly in

Address correspondence to Dr. Ralf Jockers, Department of Cell Biology, Institut Cochin, Univer-site Paris-Descartes, Faculte de Medecine, 22 rue Mechain, Paris 75014, France. E-mail: jockers@cochin.inserm.fr

Chronobiology International, 23(1&2): 419–426, (2006)Copyright # Taylor & Francis Group, LLCISSN 0742-0528 print/1525-6073 onlineDOI: 10.1080/07420520500521863

419

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

the pineal gland during the hours of darkness under the control of thesuprachiasmatic nucleus (SCN). Melatonin relays photoperiodic infor-mation by stimulating molecular targets at central and peripheral sites.Among the different melatonin binding sites described in various biologicalsystems, MT1 and MT2 receptors are the best characterized high-affinitytargets in mammals. Both receptors belong to the G protein-coupledreceptor (GPCR) super-family with seven-transmembrane–spanningdomains. GPCRs constitute the largest family of membrane receptorsand are targeted by about half of the drugs prescribed for human diseases.GPCRs are key controllers of diverse physiological processes such asneurotransmission, cellular metabolism, secretion, and cell differentiationand growth (Bockaert and Pin, 1999). Initially, GPCRs were proposed tofunction as monomers, since the heterologeous expression of a singleGPCR was usually sufficient to produce the expected pharmacology andfunction. However, data accumulated over the last 15 ys indicate thatmost, if not all, GPCRs exist as functional dimers or higher oligomericunits (Bouvier, 2001; Pin et al., 2005).

Although the precise biological significance of GPCR dimerization isstill a matter of debate, it is now well established that GPCR oligomerizationis constitutive in living cells and occurs early in the biosynthetic pathway at“physiological” receptor concentrations. Melatonin receptors also followthe general rule, since homodimers have been shown to exist for MT1

and MT2 in transfected HEK 293 cells (Ayoub et al., 2002), Xenopustectal cells (Prada et al., 2005), and chicken astrocyte cultures (Adachiet al., 2002), all of which express endogenous melatonin receptors.

After having accepted the existence of GPCR dimerization, the currentquestions in the field are rather why GPCRs need to exist as dimers or whatwould be the advantage of being a dimer? Some authors proposed thatGPCR dimerization might be necessary to pass quality-control checkpointsof the biosynthetic pathway of GPCRs (Bulenger et al., 2005). A growingnumber of observations indicate that GPCR pharmacology and functioncan be altered by heterodimerization with other members of the samereceptor super-family. This has also been shown for MT1 and MT2 thatare engaged into heterodimers when co-expressed in HEK 293 cells(Ayoub et al., 2002) (Figure 1).

Microarray-based studies confirmed the general suspicion that mul-tiple GPCRs of the 400 non-olfactory GPCRs predicted in humans areexpressed simultaneously in tissues and cells (Hakak et al., 2003). Thissuggests that most cells produce several different GPCRs that may beengaged into heterodimeric complexes if they are expressed at the sametime. A major issue for the future will be to focus on the elucidation ofthe rules governing GPCR heterodimerization versus homodimerization.The propensity of GPCRs to form dimers has only been determinedfor some receptors. Surprisingly, the most obvious hypothesis that

A. Levoye et al.420

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

homodimers show higher interaction affinities than the correspondingheterodimers turned out to be too simplistic; heterodimers appear toform with similar affinities compared to the corresponding homodimersfor most receptors studied thus far (Milligan, 2004). Interestingly, the for-mation of MT2 homodimers has been shown to be three- to four-fold lowerthan that of the MT1/MT2 heterodimer; whereas, the relative propensityof MT1 homodimer and MT1/MT2 heterodimer formation is similar(Ayoub et al., 2004), suggesting that MT2 receptors may be preferentiallyengaged into heterodimers in cells co-expressing both receptors(Figure 1).

EXISTENCE OF GPCR HETERODIMERS IN NATIVE TISSUES

Since most of the data accumulated on GPCR dimerization has beenobtained in heterologous expression systems, the existence of GPCR het-erodimers in native tissues remains an open question. For most receptors,this question has turned out to be difficult to answer. The first basic con-dition to be fulfilled would be the co-expression of the two proteins ofthe heterodimer in the same tissue, which has indeed been shown for aconsiderable number of GPCRs suspected to form heterodimers. Thebest evidence known so far for the existence of GPCR dimers comes

FIGURE 1 Homo- and heterodimerization of the MT1 and MT2 melatonin receptors. Evidence formelatonin receptor dimerization has been obtained from BRET (Bioluminescence Resonance EnergyTransfer), SDS-PAGE, and co-immunoprecipitation experiments. The Probability of dimer formationwas determined in BRET donor saturation experiments and the specific pharmacological properties ofthe heterodimer by a BRET-based assay that measures ligand-promoted conformational changes inreceptor dimers (Ayoub et al., 2002, 2004).

G Protein-Coupled Receptor Heterodimers 421

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

from co-immunoprecipitation experiments from native tissues. Colocaliza-tion studies have generally proven difficult to perform due to the lack ofadequate antibodies. Co-expression of MT1 and MT2 has been documen-ted in many melatonin-sensitive tissues, such as the hypothalamic SCN(Reppert et al., 1988), retina (Dubocovich, 1983), arteries (Krause et al.,1995), and adipose tissue (Brydon et al., 2001), suggesting that heterodi-merization could indeed occur in native mammalian tissues. In theabsence of adequate high-affinity melatonin-receptor specific antibodies,definitive proof for the existence of MT1/MT2 heterodimers cannot beprovided at the moment using co-immunoprecipitation experimentsfrom these tissues. In contrast, immunohistochemical and in-situ hybridiz-ation studies have helped to clarify this point, at least in some tissues.

Co-expression of MT1 and MT2 in pyramidal neurons of the hippo-campus, as shown by RT-PCR (Musshoff et al., 2002), is consistent withthe possible existence of MT1/MT2 heterodimers in these cells. A moredetailed data analysis showed that MT1 is predominantly expressed inthe CA1 hippocampal subfield; whereas, MT2 is predominantly expressedin the CA3/CA4 subfields (Savaskan et al., 2002b, 2005).

Expression of the mRNA of MT1 and MT2 in SCN neurons is widelyaccepted (von Gall et al., 2002). Both receptor mRNAs have been detectedin SCN neurons by in-situ hybridization (Dubocovich et al., 1998; Reppertet al., 1994). Although a detailed analysis has not been performed, theextensive labeling throughout the SCN strongly indicates co-expressionof both receptors, at least in subpopulations of SCN neurons. Detailedimmunohistochemical data are now available from the human retinawhere both, MT1 and MT2 are expressed in virtually every ganglion andphotoreceptor cell, strongly suggesting heterodimer formation (Savaskanet al., 2002b) (E.S., unpublished observations). In-situ hybridization exper-iments in cultured chicken astrocytes showed that some 25% of the cellsco-expressed the Mel1c and MT1 mRNA, which is in agreement with theformation of heterodimeric complexes (Adachi et al., 2002).

GPCR DIMERS ARE PHARMACOLOGICAL TARGETS

The ultimate aim of research on GPCR dimerization is to identify newreceptor entities with distinct pharmacological and/or signal transductionproperties to define new pharmacological targets. This aim has beenaccomplished for a growing number of GPCR heterodimers that exhibitclear differences in receptor signaling and trafficking in comparison tothe corresponding homodimers (Gazi et al., 2002; Maggio et al., 2005).Unique pharmacological properties in native tissue preparations thatcannot be explained by the pharmacology of individually expressedGPCRs have been reported in a substantial number of studies. Althoughreceptor heterodimerization is clearly only one out of several possible

A. Levoye et al.422

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

explanations for these observations, these tissues are very promising can-didates for further investigation. Increasing effort is now being extendedto identify heterodimer-specific ligands, which are ideal tools for func-tional studies in physiological settings. This aim has been accomplishedrecently for the first time with the d/k-opioid heterodimer-selectiveagonist 60-guanidinonaltrindole, confirming the existence of d/k-opioidheterodimers in the spinal cord where they mediate analgesia (Waldhoeret al., 2005). For the MT1/MT2 heterodimer, a specific pharmacologicalprofile has been established using the BRET-based assay that measuresligand-promoted conformational changes of the heterodimer (Ayoubet al., 2004). Selective ligands for the MT1/MT2 heterodimer have beenidentified, with luzindole showing the highest selectivity with a preferenceof more than 100 times for the heterodimer compared to the MT2 homo-dimer (Figure 1). Luzindole may now become a lead compound forthe development of further MT1/MT2 heterodimer-selective compoundsfor selective pharmacological intervention. Luzindole and 4-PPDOT arewidely used as MT2-selective pharmacological tools to demonstrate theimplication of this receptor subtype in specific physiological phenomena(Dubocovich et al., 1998). Since these compounds show similar or evenhigher affinity for MT1/MT2 heterodimers, the potential involvement ofthese heterodimers has to be considered when using these compoundsin cells co-expressing MT1 and MT2.

GPCR DIMERS AND PHYSIOLOGICAL FUNCTIONS

AND DISEASES

In some cases, the functional consequences of GPCR dimerization havebeen shown to be physiologically relevant (Table 1). The sense of tastedepends on the expression of proper heterodimers, with T1R1/T1R2and T1R1/T1R3 heterodimers recognizing different sweet substances(Nelson et al., 2002; Xu et al., 2004). At least part of the detection of odor-ants may also depend on the expression of the proper heterodimer pair.Despite the generally accepted rule that each olfactory neuron expressesonly one olfactory receptor, the DOR83b, a ubiquitously expressedodorant receptor in olfactory neurons of insects engages with variousother odorant receptors to form heterodimeric complexes with increasedfunctionality (Neuhaus et al., 2005).

GPCR heterodimers may also play an important role in differentdisease states. A mutant of the Frizzled 4 receptors appears to be at theorigin of the vitreoretinopathy, as this mutant traps wild-type Frizzled inthe endoplasmatic reticulum by heterodimerization (Kaykas et al., 2004).Increased expression level of the angiotensin 1 (AT1)/bradykinin 2 (B2)heterodimer plays a critical role in preeclampsia (AbdAlla et al., 2001).Future research will undoubtedly provide further evidence for the

G Protein-Coupled Receptor Heterodimers 423

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

importance of GPCR heterodimers in physiological and pathological pro-cesses and will hopefully shed light on the still unknown function of MT1/MT2 heterodimers.

ACKNOWLEDGMENTS

This work was supported by grants from the Institut National de Santeet de Recherche Medicale, the Centre National de la Recherche Scientifi-que and the Universite Paris-Descartes. AL was supported by the Fonda-tion pour la Recherche Medicale (FRM).

REFERENCES

AbdAlla, S., Lother, H., el Massiery, A., Quitterer, U. (2001). Increased AT(1) receptor heterodimers inpreeclampsia mediate enhanced angiotensin II responsiveness. Nat. Med. 7:1003–1009.

AbdAlla, S., Abdel-Baset, A., Lother, H., el Massiery, A., Quitterer, U. (2005). Mesangial AT1/B2receptor heterodimers contribute to angiotensin II hyperresponsiveness in experimentalhypertension. J. Mol. Neurosci. 26:185–192.

Adachi, A., Natesan, A.K., Whitfield-Rucker, M.G., Weigum, S.E., Cassone, V.M. (2002). Functionalmelatonin receptors and metabolic coupling in cultured chick astrocytes. Glia 39:268–278.

Ayoub, M.A., Couturier, C., Lucas-Meunier, E., Angers, S., Fossier, P., Bouvier, M., Jockers, R. (2002).Monitoring of ligand-independent dimerization and ligand-induced conformational changes ofmelatonin receptors in living cells by bioluminescence resonance energy transfer. J. Biol. Chem.277:21522–21528.

Ayoub, M.A., Levoye, A., Delagrange, P., Jockers, R. (2004). Preferential formation of MT1/MT2

melatonin receptor heterodimers with distinct ligand interaction properties compared withMT2 homodimers. Mol. Pharmacol. 66:312–321.

Bockaert, J., Pin, J.P. (1999). Molecular tinkering of G protein-coupled receptors: an evolutionarysuccess. Embo J. 18:1723–1739.

TABLE 1 Physiological Consequences of GPCR Heterodimerizationa

Heterodimer Effect of heterodimerization Reference

AT1/B2 Heterodimer plays a critical role in preeclampsia,hypersensitivity of pressor effect ofangiotensin II

(AbdAlla et al., 2001)(AbdAlla et al., 2005)

AT1/Mas Heterodimer antagonises the action ofangiotensin II

(Kostenis et al., 2005)

Fz4 (mutant)/Fz4 Trapping of wild-type Fz by vitreoretinopathy-associated Fz mutant

(Kaykas et al., 2004)

b1/b2 Optimization of cardiac contractility (Zhu et al., 2005)GABAB1/GABAB2 Heterodimerization is a prerequisite for

functional expression at the cell surface(White et al., 1998)

T1R3/T1R1T1R3/T1R2

Recognition of sweet and umami tastes dependson formation of specific heterodimers

(Nelson et al., 2001)(Nelson et al., 2002)(Xu et al., 2004)

DOR83b/DOR43a Improved receptor functionality (Neuhaus et al., 2005)

aAT1, type1 angiotensin II receptor; b1(2), b1/(2)-adrenergic receptor; B2, bradikinin receptor 2;DOR83b(43a), Drosophila odorant receptor 83b(43a); Fz4, Frizzled 4 receptor; GABAB1(2), metabotro-pic g-aminobutyric acid (GABA) B receptor 1(2); Mas, mas proto-oncogene; T1R, T1R taste receptor.

A. Levoye et al.424

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

Bouvier, M. (2001). Oligomerization of G protein-coupled transmitter receptors. Nat. Rev. Neurosci.2:274–286.

Brydon, L., Petit, L., Delagrange, P., Strosberg, A.D., Jockers, R. (2001). Functional expression of MT2

(mel1b) melatonin receptors in human paz6 adipocytes. Endocrinology 142:4264–4271.Bulenger, S., Marullo, S., Bouvier, M. (2005). Emerging role of homo- and heterodimerization in

G protein-coupled receptor biosynthesis and maturation. Trends Pharmacol. Sci. 26:131–137.Dubocovich, M.L. (1983). Melatonin is a potent modulator of dopamine release in the retina. Nature

306:782–784.Dubocovich, M.L., Yun, K., AlGhoul, W.M., Benloucif, S., Masana, M.I. (1998). Selective MT2

melatonin receptor antagonists block melatonin-mediated phase advances of circadianrhythms. FASEB J. 12:1211–1220.

Gazi, L., Lopez-Gimenez, J.F., Strange, P.G. (2002). Formation of oligomers by G protein-coupledreceptors. Curr. Opin. Drug Discov. Devel. 5:756–763.

Hakak, Y., Shrestha, D., Goegel, M.C., Behan, D.P., Chalmers, D.T. (2003). Global analysis of Gprotein-coupled receptor signaling in human tissues. FEBS Lett. 550:11–17.

Kaykas, A., Yang-Snyder, J., Heroux, M., Shah, K.V., Bouvier, M., Moon, R.T. (2004). Mutant Frizzled4 associated with vitreoretinopathy traps wild-type Frizzled in the endoplasmic reticulum byoligomerization. Nat. Cell Biol. 6:52–58.

Kostenis, E., Milligan, G., Christopoulos, A., Sanchez-Ferrer, C.F., Heringer-Walther, S., Sexton, P.M.,Gembardt, F., Kellett, E., Martini, L., Vanderheyden, P., Schultheiss, H.P., Walther, T. (2005).G protein-coupled receptor Mas is a physiological antagonist of the angiotensin II type 1 receptor.Circulation 111:1806–1813.

Krause, D.N., Barrios, V.E., Duckles, S.P. (1995). Melatonin receptors mediate potentiation ofcontractile responses to adrenergic nerve stimulation in rat caudal artery. Eur. J. Pharmacol.276:207–213.

Maggio, R., Novi, F., Scarselli, M., Corsini, G.U. (2005). The impact of G protein-coupled receptorhetero-oligomerization on function and pharmacology. Febs J. 272:2939–2946.

Milligan, G. (2004). G protein-coupled receptor dimerization: function and ligand pharmacology. Mol.Pharmacol. 66:1–7.

Musshoff, U., Riewenherm, D., Berger, E., Fauteck, J.D., Speckmann, E.J. (2002). Melatonin receptorsin rat hippocampus: Molecular and functional investigations. Hippocampus 12:165–173.

Nelson, G., Hoon, M.A., Chandrashekar, J., Zhang, Y., Ryba, N.J., Zuker, C.S. (2001). Mammaliansweet taste receptors. Cell 106:381–390.

Nelson, G., Chandrashekar, J., Hoon, M.A., Feng, L., Zhao, G., Ryba, N.J., Zuker, C.S. (2002). Anamino-acid taste receptor. Nature 416:199–202.

Neuhaus, E.M., Gisselmann, G., Zhang, W., Dooley, R., Stortkuhl, K., Hatt, H. (2005). Odorantreceptor heterodimerization in the olfactory system of Drosophila melanogaster. Nat. Neurosci.8:15–17.

Pin, J.P., Kniazeff, J., Liu, J., Binet, V., Goudet, C., Rondard, P., Prezeau, L. (2005). Allosteric function-ing of dimeric class C G protein-coupled receptors. Febs J. 272:2947–2955.

Prada, C., Udin, S.B., Wiechmann, A.F., Zhdanova, I.V. (2005). Stimulation of melatonin receptorsdecreases calcium levels in xenopus tectal cells by activating GABA(C) receptors. J. Neurophysiol.94:968–978.

Reppert, S.M., Weaver, D.R., Rivkees, S.A., Stopa, E.G. (1988). Putative melatonin receptors in ahuman biological clock. Science 242:78–81.

Reppert, S.M., Weaver, D.R., Ebisawa, T. (1994). Cloning and characterization of a mammalianmelatonin receptor that mediates reproductive and circadian responses. Neuron 13:1177–1185.

Savaskan, E., Olivieri, G., Meier, F., Brydon, L., Jockers, R., Ravid, R., Wirz-Justice, A., Muller-Spahn, F. (2002a). Increased melatonin 1a-receptor immunoreactivity in the hippocampus ofAlzheimer’s disease patients. J. Pineal. Res. 32:59–62.

Savaskan, E., Wirz-Justice, A., Olivieri, G., Pache, M., Krauchi, K., Brydon, L., Jockers, R., Muller-Spahn, F., Meyer, P. (2002b). Distribution of melatonin MT1 receptor immunoreactivity inhuman retina. J. Histochem. Cytochem. 50:519–526.

Savaskan, E., Ayoub, M.A., Ravid, R., Angeloni, D., Fraschini, F., Meier, F., Eckert, A., Muller-Spahn, F., Jockers, R. (2005). Reduced hippocampal MT2 melatonin receptor expression inAlzheimer’s disease. J. Pineal. Res. 38:10–16.

G Protein-Coupled Receptor Heterodimers 425

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.

von Gall, C., Stehle, J.H., Weaver, D.R. (2002). Mammalian melatonin receptors: molecular biologyand signal transduction. Cell Tissue Res. 309:151–162.

Waldhoer, M., Fong, J., Jones, R.M., Lunzer, M.M., Sharma, S.K., Kostenis, E., Portoghese, P.S.,Whistler, J.L. (2005). A heterodimer-selective agonist shows in vivo relevance of G protein-coupled receptor dimers. Proc. Natl. Acad. Sci. USA 102:9050–9055.

White, J.H., Wise, A., Main, M.J., Green, A., Fraser, N.J., Disney, G.H., Barnes, A.A., Emson, P.,Foord, S.M., Marshall, F.H. (1998). Heterodimerization is required for the formation of afunctional GABA(B) receptor. Nature 396:679–682.

Xu, H., Staszewski, L., Tang, H., Adler, E., Zoller, M., Li, X. (2004). Different functional roles of T1Rsubunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. USA 101:14258–14263.

Zhu, W.Z., Chakir, K., Zhang, S., Yang, D., Lavoie, C., Bouvier, M., Hebert, T.E., Lakatta, E.G.,Cheng, H., Xiao, R.P. (2005). Heterodimerization of beta1- and beta2-adrenergic receptorsubtypes optimizes beta-adrenergic modulation of cardiac contractility. Circ. Res. 97:244–251.

A. Levoye et al.426

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y A

thlo

ne I

nstit

ute

of T

echn

olog

y on

08/

20/1

3Fo

r pe

rson

al u

se o

nly.