N EUROPSYCHOPHARMACOLOGY 2000 VOL 23 NO S4 2000 America - PDF document

1 N EUROPSYCHOPHARMACOLOGY 2000 Ð VOL .
N EUROPSYCHOPHARMACOLOGY 2000 Ð VOL . 23 , NO . S4 © 2000 American College of Neuropsychopharmacology Evidence for Adenosine/Dopamine Receptor Interactions: Indications for Heteromerization Rafael Franco, Ph.D., Sergi FerrŽ, M.D., Ph.D., Luigi Agnati, Ph.D., Maria Torvinen, B.A.,Silvia GinŽs, B.A., Jo‘lle Hillion, Ph.D., Vicent Casad—, Ph.D., Pierre-Marie Lled—, Ph.D., Michele Zoli, Ph.D., Carmen Lluis, Ph.D., and Kjell Fuxe, M.D., Ph.D. Evidence has been obtained for adenosine/dopamine interactions in the central nervous system. There exists an anatomical basis for the existence of functional interactions between adenosine A KEY WORDS : N EUROPSYCHOPHARMACOLOGY 2000 Ð VOL . 23 , NO . S4 Adenosine/Dopamine Receptor Interactions S51 ganized in Stockholm by Agnati and Fuxe (Fuxe andAgnati 1987). G-protein-coupled receptors are often re-garded as transmembrane allosteric proteins of a mono-meric character (see Changeux and Edelstein 1998). Butalready in 1982 the ordered formation and stabilizationof GPCR clusters called receptor mosaics were postu-lated, as well as their participation in learning andmemory (Agnati et al. 1982).Dimerization of membrane receptors was first dem-onstrated for the tyrosine kinase receptor superfamily,a phenomenon which appeared to be essential for sig-nal transduction including autophosphorylation andenhancement of affinity for the agonist (Schlessinger1988; Schlessinger and Ullrich 1992). Heterodimeriza-tion was also clearly demonstrated for the tyrosine ki-nase receptors (Ullrich and Schlessinger 1990). In the1980s there also appeared indications that functionalG-protein-linked receptors exist in a dimeric form (Venterand Fraser 1983; Conn et al. 1982). Evidence for direct mo-lecular crosstalk between GPCR was indeed obtained byMaggio and colleagues using chimeric muscarinic/adren-ergic receptors (Maggio et al. 1993). Therefore, we postu-lated in 1993 that several cases of G-protein-coupled re-ceptorÐreceptor interactions occurring in crude membranepreparations may be based on a process of heterodimer-ization, because the molecular mechanisms for dimeriza-tion may be conserved among different subclasses ofGPCR (Zoli et al. 1993). The density of two or more inter-acting receptors and the number of receptors activated bythe agonist in each population would then determine theproportion of monomers and homodimers and het-erodimers and thus the overall action on target cellfunction (Zoli et al. 1993).At that time the first evidence of the existence ofGPCR dimers was obtained by Ng et al. (1993, 1994a,1994b, 1996) using antibodies specific for GPCR. It wasshown in 1993 that the 5HT-1B receptor exists as mono-mers and dimers (Ng et al. 1993). This was followed bydemonstration of dimers and oligomers of D 1 and D 2 re-ceptors in infected Sf cells (Ng et al. 1994a, 1994b, 1996)and of A 1 receptors in a natural cell line and in mamma-lian brain (Ciruela et al. 1995). More recently, direct evi-dence for GPCR heterodimerization and higher orderoligomerization (Figure 1) was obtained for GABA B(see Jones et al. 1999) and of kappa and delta opioid re-ceptors (Jordan and Devi 1999). This opens a new fieldof drug development where molecular models ofhomo- and heterodimers will play an important role. Atthe same time the possible existence of additional mecha-nisms for receptorÐreceptor interaction must be consid-ered, involving so-called anchoring (adaptor) proteins(see, e.g., Homers in the case of the metabotropic gluta-mate receptor 1 and 5), which can play a role in forma-tion of heteromeric (and may be homomeric) complexes(Brakeman et al. 1997) (Figure 1). The G-protein net-work involving the receptor crosstalk through G-pro-tein beta-gamma release and exchange should also beconsidered (see Zoli e

2 t al. 1993). DOPAMINE-ADENOSINE INTERACT
t al. 1993). DOPAMINE-ADENOSINE INTERACTIONS IN THE CENTRAL NERVOUS SYSTEM Adenosine is an endogenous nucleoside acting as aneuromodulator in the central nervous system. Its ac-tions are mediated by adenosine receptors, four ofwhich have been cloned and pharmacologically charac-terized: A 1 , A 2A , A 2B and A 3 (Fredholm et al. 1994).Among those four subtypes A 1 and A 2A are the maintargets of the behavioral effects occurring in animalstreated with adenosine analogs (FerrŽ et al. 1992, 1993,1997; Fredholm 1995). Caffeine, as an example of an an-tagonist acting at adenosine receptors, is today the mostconsumed psychostimulant drug in the world.A 1 and A 2A receptors localized in the basal ganglia,and more precisely in the striatum, are responsible forthe motor depressant effects of adenosine agonists andof the motor stimulatory effects of adenosine antago-nists (FerrŽ et al. 1992, 1997). A vast majority of striataladenosine receptors are located in the medium-sizedspiny GABAergic neurons, efferent neurons that consti-tute more than 90% of the neuronal population in thestriatum (Schiffmann et al. 1991; Rivkees et al. 1995).One subtype of GABAergic efferent neurons, the strio-pallidal ones, are those mainly containing D 2 dopamin-ergic receptors. A second subtype, the strionigro-stri-oentopeduncular neurons, contains D 1 dopaminergicreceptors (see FerrŽ et al. 1997). The two subtypes ofneurons both contain A 1 adenosine receptors, but onlythe striopallidal neurons contain A 2A receptors. Theseanatomical localizations indicate that in the basal gan-glia A 2A receptors only exist in the D 2 receptor-contain-ing neurons, whereas A 1 receptors exist in both D 1 andD 2 receptor-containing nerve cells. The anatomical lo-calization of dopamine and adenosine receptor subtypeshave provided the anatomical basis for the existence offunctional interaction between A 1 and D 1 or between A 2A and D 2 receptors in the same neurons. These functional in-teractions have been investigated using a variety of tech-niques, from ligand binding and second messenger deter-minations to behavioral studies.In terms of ligand binding the presence of adenosineanalogues modify the affinity of dopamine analoguesfor the binding to D 1 receptors. Selective A 1 R agonistsaffect negatively the high-affinity binding of dopamineto D 1 receptors (FerrŽ et al. 1998). These results havebeen obtained working with membranes from tissues orfrom cells cotransfected with the cDNA for the humanversions of A 1 R and D 1 R. The adenosine A 1 agonistsshift the high-affinity binding to the low-affinity statesof the D 1 receptor. On the other hand, there is also an in-teraction at the adenylate cyclase level. A 1 R are nega- S52 R. Franco et al.N EUROPSYCHOPHARMACOLOGY 2000 Ð VOL . 23 , NO . S4 tively coupled to the adenylate cyclase whereas D 1 re-ceptors are positively coupled to this enzyme. Thus, thepresence of adenosine agonists leads to a decrease in thedopamine D 1 -induced cAMP production whereas, inturn, the presence of antagonists acting on A 1 R leads tothe potentiation in the cAMP responses occurring by ac-tivation of D 1 receptors (FerrŽ et al. 1998).Adenosine agonists acting upon A 2A receptors in-stead counteract the effect of dopamine acting uponD 2 R. In fact, activation of A 2A receptors leads to a de-crease in receptor affinity for dopamine agonists, espe-cially of the high- affinity state (FerrŽ et al. 1991; Das-gupta et al. 1996; Kull et al. 1999). Even in cotransfectedCHO cells treatment with the A 2A selective agonistCGS21680 leads to a marked reduction in the binding ofdopamine to D 2 receptors (Figure 2). Recently, in a neu-roblastoma cell line, which constitutively expressesA 2A R, it has been shown that transfection of D 2 R lead

3 sto an antagonistic adenosine/dopamine c
sto an antagonistic adenosine/dopamine crosstalkwhich is mediated by A 2A and D 2 receptor interactions.This functional interaction is demonstrated to occur inacute treatments in which the dopamine-induced coun-teraction of the increase in intracellular calcium concen-tration evoked by KCl is blocked via simultaneous acti-vation of A 2A R (Salim et al. 2000).Adenosine/dopamine interactions at the behaviorallevel probably reflect those found at the level of recep-tor binding and signaling. Thus, adenosine receptor an-tagonist-induced motor activation is counteracted bytreatments that cause an acute dopamine depletion orblockade of D 1 or D 2 receptors (FerrŽ et al. 1992, 1997).Furthermore, adenosine receptor agonists inhibit andadenosine receptor antagonists potentiate the motor ac-tivating effects of dopamine agonists (FerrŽ et al. 1992,1997). Specifically, low doses of A 1 and A 2A receptor ag-onists selectively counteract the motor activating effectsinduced by D 1 and D 2 receptor agonists, respectively.On the other hand, selective A 1 R antagonists selectivelypotentiate D 1 R receptor agonist-induced motor activa-tion whereas A 2A receptor antagonists potentiate D 2 re-ceptor agonist-mediated motor effects (see FerrŽ et al.1997 for review). Figure 1.Different types of receptorÐreceptor interactions. The so-called ÒdirectÓ receptorÐreceptor interactions would beexemplified by receptor heterodimerization. ÒQuasi directÓ receptorÐreceptor interactions would involve anchoring pro-teins (such as Homers in the case of metabotropic glutamate receptors). N EUROPSYCHOPHARMACOLOGY 2000 Ð VOL . 23 , NO . S4 Adenosine/Dopamine Receptor Interactions S53 Altogether, the correlation among the data obtainedat the cellular level, at the behavioral level and even atthe level of neuronal function (see FerrŽ et al. 1997)strongly suggest that the receptor subtype-specific in-teraction between adenosine and dopamine receptors(A 2A /D 2 and A 1 /D 1 ) play an essential role in the modu-lation of basal ganglia function. To some extent, theseinteractions are a consequence of specific molecular in-teractions achieved by means of heteromerization (seebelow). ADENOSINE RECEPTORS AS A MODEL FOR HOMOMERIC AND HETEROMERICPROTEINÐPROTEIN INTERACTIONS The work of Franco and colleagues on A 1 adenosine re-ceptors has provided a better understanding of howligand binding and signal transduction is affected byhomotypic and heterotypic proteinÐprotein interactions(Franco et al. 1996, 1997; GinŽs et al. 2000; Sarri— et al.2000).The binding of [ 3 H]-2-chloroadenosine to A 1 recep-tors present in rat brain membranes was first studied intwo laboratories. Basically, depending upon the ab-sence or presence of exogenous adenosine deaminase(ADA), one single (low-affinity; Wu et al. 1980; Wu andPhillis 1982) or two binding sites (low- and high-affin-ity; Williams and Risley 1980a, 1980b) were found. Theappearance of a high-affinity binding site in the pres-ence of ADA was explained by the disappearance of en-dogenous adenosine, which acts as a competitor of A 1 Ragonists, or by assuming that ADA had an extracata-lytic high-affinity binding site for 2-chloroadenosine(Phillis and Wu 1981), which is not the case accordingto the X-ray structure of the enzyme (Wilson et al.1991). Subsequent studies have demonstrated that A 1 Rpresent two different affinities for agonists that dependon the coupling to heterotrimeric G-proteins (Lohse etal. 1984); coupled receptor G-protein complexes displayhigh affinity for agonists (K d 5 0.1Ð0.2 nM), whereas un-coupled receptors display low affinity (1Ð2 nM) (Lohseet al. 1984; Casad— et al. 1990).Although for many years it has been considered thatexogenous ADA acts by removing endogenous adenos-ine, this may not have been the true

4 explanation for therevelation of a high
explanation for therevelation of a high-affinity binding site. FrancoÕs labora-tory has accumulated sufficient evidence to be sure thatADA and A 1 R interact and that both proteins are func-tionally coupled. First, in pig brain cortex membraneswhere the adenosine concentration was undetectable, itwas found that ADA was necessary for the identification Figure 2.Representative saturation curves of specific binding of [3H]dopamine ([7,8-3H]dopamine; 40 Ci/mmol; Amer-sham Pharmacia) in membrane preparations from CHO cells stably cotransfected with human adenosine A2A and ratdopamine D2S receptor cDNAs (for details about the transfection and maintenance of the cells, as well as the membranepreparation, see Kull et al. 1999) in the presence and absence of the adenosine A2A agonist CGS 21680 (100 nM). Apomor-phine (0.2 mM) was used for nonspecific binding. Results represent means 6 standard deviation of triplicate data of one sin-gle experiment. Bmax and KD values were 382 fmol/mg prot. and 3.3 nM, respectively, for the control curve and 578 fmol/mgprot. and 15.3 nM, respectively, in the presence of CGS 21680. The results show a predominant modulatory effect of adenos-ine A2A receptors on the affinity of dopamine D2S receptors (a five-fold decrease), since [3H]dopamine labels the high-affinitycomponent of the transfected dopamine D2S receptors, at the concentrations used in the present experiment (original data). S54 R. Franco et al.N EUROPSYCHOPHARMACOLOGY 2000 Ð VOL . 23 , NO . S4 of the high-affinity component of the binding to A 1 R. Infact, in the absence of ADA, a single low-affinity bind-ing was found (Table 1). If the high-affinity site corre-sponds to the receptor G-protein complex, ADA wouldbe necessary for the coupling of A 1 R to G-proteins. How-ever, the cluster-arranged cooperative model, which ac-counts for the kinetics of ligand binding to A 1 R (Franco etal. 1996), shows that high- and low-affinity sites are a con-sequence of the negative cooperativity of agonist bindingand may not be related to the content of free and GPCR.Therefore, ADA would affect cooperativity without af-fecting the A 1 R-G protein coupling. Further investiga-tions of the molecular interaction between ADA and A 1 Rwas performed in a smooth muscle cell line (DDT 1 MF-2),which is currently used as a model of A 1 R-expressingcells. Furthermore, the expressed A 1 R display similarbinding kinetics as A 1 R present in cerebral cortex. Inthese cells, apart from the confirmation of an increase inthe affinity for agonists only in the presence of ADA, amore molecular approach was used (Ciruela et al. 1996).Thus, the proof of an ADA/A 1 R interaction was investi-gated by means of confocal microscopy, affinity chroma-tography and coimmunoprecipitation. Some of these ex-periments were made possible by the development ofantibodies against A 1 R which worked well in immunocy-tochemical, immunoprecipitation and immunoblottingassays. All the approaches tried were positive sinceADA and A 1 R coprecipitated, A1R was specifically re-tained in a matrix of ADA-Sepharose and both ADAand A1R colocalized on the surface of DDT1MF-2 cells.In these cells, the degree of colocalization betweenADA and CD26, an ADA-anchoring protein in lympho-cytes, was lower than the colocalization between ADAand A1R. Interestingly, when cells were preincubatedwith commercial ADA (from calf intestine), the degreeof colocalization ADA/A1R approached 100%.These data constituted the first evidence demonstrat-ing an interaction between a degradative ectoenzymeand the receptor whose ligand is the enzyme substrate.Due to the fact that the interaction of ADA with CD26on the surface of lymphocytes leads to signal transduc-tion it was suspected that the interaction ADA/A1Rmight have a role in the modulati

5 on of the receptor-me-diated responses.
on of the receptor-me-diated responses. Using a compound able to inhibit theenzymatic activity it was demonstrated that the ADA/A1R interaction was needed for an efficient signalingvia the adenosine receptor. Therefore, ADA is actingcatalytically but also in an extraenzymatic way as a co-stimulatory molecule. The actual scenario is that at lowadenosine concentrations ADA is mainly facilitatingsignaling whereas at high adenosine concentrations, bywhich the ADA/A1R interaction is disrupted, the lowaffinity state of the receptor predominates and an autol-ogous mechanism of desensitization is devised (seeFranco et al. 1997 for review).Apart from this heterotypic interaction A1R do formhomodimers in membranes from a variety of tissues orcell lines. The first evidence was obtained by immunob-lotting using antibodies that recognized, in samplesfrom pig brain cortical membranes, a specific band of 39kDa and, in addition, a second band of 74 kDa (Figure3). This high molecular weight band did not dissociatein a reducing environment or by the treatment at 1008with detergent and did not contain G-proteins thatcould be forming a stable complex with the receptor(Figure 3). Therefore the band probably reflected the ex-istence of dimers in membranes from pig brain. Thebands corresponding to the monomer and to the dimerwere present in extracts from different pig and rat tis-sues, the dimer being specially abundant in samplesfrom cortex and striatum.A1R, like other GPCR are multifunctional proteinswhich are able to form high- order molecular structurescontaining at least two receptor molecules, heterotrim-eric G-proteins and adenosine deaminase. The search ofother proteins interacting with A1R are underway andat least one more protein able to interact with an intrac-ellular loop of those receptors has been discovered (Sar-ri— et al. 2000). This protein, hsc73, affects the binding ofadenosine deaminase to A1R and the preliminary re-Table 1.Equilibrium Parameters of [3H]R-PIA Binding to DDT1MF-2 Cells and to Cell Membranes in the Absence or Presence of ADAPresence of ADAAffinity StateKd (nM)Bmax (pmol/mg prot)NoneHigh-affinityÐÐLow-affinityÐÐVery low-affinity50 6 100.4 6 0.10.2 U/mlHigh-affinity0.79 6 0.090.28 6 0.03Low-affinity9 6 20.15 6 0.05Very low-affinityÐÐ0-2 U/ml plus Hg21High-affinity1.5 6 0.50.25 6 0.05Low-affinity9 6 20.13 6 0.07Very low-affinityÐÐHg21 was used to abolish the enzymatic activity of ADA without disrupting the interaction with A1 aden-osine receptor (modified from Ciruela et al. 1996). NEUROPSYCHOPHARMACOLOGY 2000ÐVOL. 23, NO. S4Adenosine/Dopamine Receptor InteractionsS55 sults indicate that A1R cannot bind to both proteins atthe same time. It should be noted that all these interac-tions play a role in ligand binding and in signaling butalso in traffic and down-regulation of the receptors.DIMERS AND CLUSTERS OF GPCRsSince the description of the existence of homodimers for5HT1B, D1, D2 (Ng et al. 1993, 1994a, 1994b, 1995) andA1 receptors (Ciruela et al. 1995) a number of reportshave described the occurrence of homodimers for a vari-ety of GPCRs. In fact it now seems that any member ofthe GPCR superfamily can be present in form of dimersin the plasma membrane. To our knowledge reports onthe existence of dimers have been disclosed for dopa-mine (Ng et al. 1993, 1994a, 1994b, 1995; Zawarynski etal. 1998), adrenergic (Hebert et al. 1996), metabotropicglutamate (Romano et al. 1996), serotonin (Pauwels et al.1998; Xie et al. 1999), muscarinic acetylcholine (Maggioet al. 1996; Zeng and Wess 1999), opioid (Cvejic andDevi 1997) and chemokine (Rodriguez-Frade et al. 1999)receptors.In view of the recent discovery of dimers for GPCRthere are efforts to know the molecular mechanism of theinteraction. Although an indirect interaction through abridg

6 e protein cannot be discarded, the evide
e protein cannot be discarded, the evidence so farindicates that some transmembrane domains participatein the interaction and that the interaction involves mainlynonpolar residues. The use of theoretical biology ap-proaches by Gouldson and colleagues (Gouldson et al.1997, 1998) have predicted that proteinÐprotein interac-tion for the homodimer formation (and eventually of het-erodimers) is mediated by domain swapping involvingtransmembrane regions 5 and 6, and probably other he-lical transmembrane domains as well. The only excep-tion to this general trend are metabotropic glutamate 5receptors for which a disulfide-linked structure hasbeen demonstrated for the dimer (Romano et al. 1996).This may be rather unique to these receptors since theyshow little homology with other members of the GPCRfamily.Assuming that there is an equilibrium between mono-meric and heteromeric forms of GPCR in membranes, theagonists displace the equilibrium toward the formation ofthe dimer. It seems, at least for some members of theGPCR family, that this displacement occurs becausedimers are more functional that monomers. But interest-ingly, when observed under a microscope in immunocy-tochemical assays, adenosine receptors cluster in the pres-ence of agonists (Figure 4). It is highly probable thatclustering occurs for several GPCRs when they are acti-vated by their ligands. Occurrence of clustering clearly re-flects that GPCR form high molecular order structures inwhich multimers of the receptors and, probably, other in-teracting proteins form functional complexes whose pre-cise role in the biochemistry and physiology of the recep-tor will require more experimental effort in the future.DOPAMINE-ADENOSINE RECEPTORÐRECEPTOR INTERACTIONSSome of the functional interactions between dopamineand adenosine can be explained by downstream inter-actions. Thus the A1/D1 receptor antagonism at thelevel of the cAMP formation can be easily explained bythe fact that G-proteins for A1R and D1R are differently Figure 3.Immunoblotting analysis of the pig brain adenosine A1R. Pig cortical brain membranes were photoaffinitylabelled with [125I]R-AHPIA in the absence (A) or in the presence (B) of an excess of unlabelled R-AHPIA. Seventy micro-grams of protein from cortical pig brain crude membranes (lane 1) or detergent extracts (lanes 2Ð6) were used. Detergentextracts were prepared from untreated membranes (lane 2) or from membranes treated with either R-AHPIA (lane 4),DPCPX (lane 5) or R-PIA plus Gpp(NH)p (lane 6). In lane 3 the acetone precipitated of detergent extracts was applied. Aftertransferring and blotting, the PVDF membrane was incubated with two anti-A1R antibodies: PC10 (against an intracellularloop) and PC20 (against an extracellular loop), with antibodies anti G-alpha, or with nonimmune serum. Arrows: bands cor-responding to monomers and dimers of A1R. Arrowhead: band corresponding to G-alpha subunits (from Ciruela et al. 1995,with permission). S56R. Franco et al.NEUROPSYCHOPHARMACOLOGY 2000ÐVOL. 23, NO. S4 coupled to adenylate cyclase. However, some of thefunctional interactions occur even in acute treatments(see above) which suggested that a direct interactionmay occur. Due to the occurrence of homodimers of ad-enosine and of dopamine receptors it is tempting tospeculate that heterodimers between receptors belong-ing to two different families of GPCR can be formed.Until recently it had not been possible to study this pos-sibility due to the lack of suitable tools in the form ofspecific antibodies working for immunoprecipitation,immunoblotting and immunofluorescence studies. Wehave now enough evidence to be sure that A1 and D1form heterodimers in both cotransfected cells and inprimary cultures of neurons. In fact it has been possibleto see a high colocalization between

7 A1R and D1R evenin primary cultures and
A1R and D1R evenin primary cultures and also antibodies against the A1Rare able to coimmunoprecipitate the D1R (GinŽs et al.2000). A further proof of the specificity of the interac-tion has been obtained by studying the effect of ligandson the distribution of the receptors in the membrane.Thus, whereas agonists for A1R cluster the two recep-tors in cotransfected cells, the ligand for D1R clusterD1R but not A1R and, therefore, the interaction is lost(GinŽs et al. 2000). It is rewarding to have been able toprove our hypothesis emitted in the 1980s and note thatligands can modulate the distribution of the receptorsin the membrane in what we can call the dancing of re-ceptors.Similar experiments are currently underway to try todemonstrate the existence of such interaction betweenA2A and D2 receptors. While we are raising antibodiesuseful for coimmunoprecipitation experiments we haveobtained excellent results in terms of colocalization be-tween A2A and D2 receptors in cell lines and in primarycultures from striatum. Hopefully we will be able todemonstrate soon that the two receptors coimmunopre-cipitate.HOMO- AND HETEROMERIZATION OF GPCR: UNDERSTANDING THE NERVOUS SYSTEMIn 1999 the first reports on heteromerization involvingGPCR started to appear in the literature. They corre-sponded to heterodimerization among receptors for thesame ligand. Heterodimers consisting of two subtypesof opioid receptors (k and d) have a pharmacologicalprofile that differs from that corresponding to each ofthe receptor subtypes when expressed alone (Jordanand Devi 1999). The case of heterodimerization ofGABABR1 and GABABR2 receptors is paradigmaticsince cells only express these receptors when they areassembled together in the endoplasmic reticulum(Jones et al. 1999; Kaupmann et al. 1999; White et al.1999). An interaction between a GPCR for dopamine(D5 receptors) and a ligand gated receptor for GABA(GABAA receptors) has just been reported (Liu et al.2000). Also our report on heteromerization of twoGPCR receptors for two structurally different ligands(A1R and D1R) has just appeared (GinŽs et al. 2000).Taken together the existence of homodimers and het-erodimers of various types indicates that the operationof those receptors may involve homotypic and hetero-typic interactions which are crucial for GPCR functionand for ligand-gated channels. It is quite reasonablethat the interacting proteins might assemble and disas-semble depending upon the composition of the cellmembrane (i.e., the types of receptors present there),and upon the presence of the different neuromodula- Figure 4.Ligand-induced clustering of A1R in DDT1-MF2cells. Anti-A1R antibodies was used to stain control cells (A) orcells treated with 50 nM R-PIA for 5 min (B). Scale bar: 10 mm. NEUROPSYCHOPHARMACOLOGY 2000ÐVOL. 23, NO. S4Adenosine/Dopamine Receptor InteractionsS57 tors in the extracellular medium. As an example, there isevidence that A1R interact with metabotropic glutamatereceptors in cerebellar neurons (Ciruela et al., data inpreparation). Therefore, the heterotypic interactions es-tablished with A1R in a given cell would depend uponthe presence of D1R, of metabotropic glutamate receptorsor both. Furthermore, the geometry of the interactionsin the clusters formed after activation of the receptorsmay depend upon the combination of neuromodulatorspresent in the extracellular medium, and more pre-cisely in the synapses.In the central nervous system this scenario is very at-tractive because receptorÐreceptor interactions arelikely involved in neurotransmission and neuromodu-lation as well as in development or in the establishmentof higher neural functions such learning and memory(Agnati et al. 1982). At this point you are reminded ofthe model of the fluid mosaic of receptors that we de-vised many years

8 ago. We think that at a given plasmamem
ago. We think that at a given plasmamembrane a multiple interaction between receptorsand adaptor proteins, such as ADA itself, exists (Figure1). This leads to a mosaic which is specific for a givenneuron and is responsible for a given action of, say, agiven neuromodulator. On the other hand, this mosaicis dynamic because ligands might affect the composi-tion and the geometry of the mosaic. This may consti-tute the basis for a better understanding of how the cen-tral nervous communication network is supported bythe multifunctional role of neuronal GPCR.REFERENCESAgnati L, Fuxe K, Zoli M, Rondanini C, Ogren SO (1982):New vistas on synaptic plasticity: mosaic hypothesis ofthe engram. Med Biol 60:183Ð190Agnati L, Benfenati F, Solfrini V, Biagini G, Fuxe K, GuidolinD, Carani C, Zini I (1993): Intramembrane receptorÐrecep-tor interactions: intergration of signal transduction path-ways in the nervous system. Neurochem Int 22:213Ð222Brakeman PR, Lanahan AA, OÕBrien R, Roche K, Barnes CA,Huganir RL, Worley PF (1997): Homer: a protein thatselectively binds metabotropic glutamate receptors.Nature 386:284Ð288Casad— V, Cant’ C, Mallol J, Canela EI, Lluis C, Franco R(1990): Solubilization of A1 adenosine receptor from pigbrain: characterization and evidence of the role of thecell membrane on the coexistence of high- and low-affinity states. J Neurosc Res 26:461Ð473Changeux J-P, Edelstein SJ (1998): Allosteric receptors after30 years. Neuron 21:959Ð980Ciruela F, Casad— V, Mallol J, Canela EI, Lluis C, Franco R(1995): Immunological identification of A1 adenosinereceptors in brain cortex. J Neurosc Res 42:818Ð828Ciruela F, Saura C, Canela EI, Mallol J, Lluis C, Franco R(1996): Adenosine deaminase affects ligand-inducedsignalling by interacting with cell surface adenosinereceptors. FEBS Lett 380:219Ð223Conn PM, Rogers DC, McNeil R. (1982) Potency enhance-ment of a GnRH agonist: GnRH-receptor microaggrega-tion stimulates gonadotropin release. Endocrinology111:335Ð337Cvejic S, Devi LA (1997): Dimerization of the delta opioidreceptor: implication for a role in receptor internaliza-tion. J Biol Chem 272:26959Ð26964Dasgupta S, FerrŽ S, Kull B, Hedlund P, Finnman V, AhlbergS, Arenas E, Fredholm BB, Fuxe K (1996): AdenosineA2A receptors modulate the binding characteristics ofdopamine D2 receptors in stably cotransfected fibro-blast cells. Eur J Pharmacol 316:325Ð331FerrŽ S, von Euler G, Johansson B, Fredholm BB, Fuxe K(1991): Stimulation of high affinity adenosine A-2 recep-tors decreases the affinity of dopamine D-2 receptors inrat striatal membranes. Proc Natl Acad Sci USA88:7238Ð7241FerrŽ S, Fuxe K, von Euler G, Johansson B, Fredholm BB(1992): Adenosine-dopamine interactions in the brain.Neuroscience 51:501Ð512FerrŽ S, OÕConnor WT, Fuxe K, Ugerstedt U (1993): Thestriopallidal neuron: a main locus for adenosine-dopamine interactions in the brain. J Neurosc 13:5402Ð5406FerrŽ S, Fredholm BB, Morelli M, Popoli P, Fuxe K (1997):Adenosine-dopmaine interactions as an integrativemechanism in the basal ganglia. Trends Neurosc20:482Ð487FerrŽ S, Torvinen M, Antoniou K, Irenius E, Civelli O, Are-nas E, Fredholm BB, Fuxe K (1998): Adenosine A1receptor-medaited modulation of dopamine D1 recep-tors in stably cotransfected fibroblasts cells. J Biol Chem273:4718Ð4724Franco R, Casad— V, Ciruela F, Mallol J, Lluis C, Canela EI(1996): The cluster-arranged cooperative model: amodel that accounts for the binding kinetics to A1 ade-nosine receptors. Biochemistry 35:3007Ð3015Franco R, Casad— V, Ciruela F, Saura C, Mallol J, Canela EI,Lluis C (1997): Cell surface adenosine deaminase: muchmore than an ectoenzyme. Prog Neurobiol 52:283Ð294Fredholm BB, Abbracchio MP, Burnstock G, Daly JW,Harden TK, Jacobson KA, Leff P, Williams M (1994):Nomenclature and classification of purinoceptors.

9 Phar-macol Rev 46:143Ð156Fredholm BB (19
Phar-macol Rev 46:143Ð156Fredholm BB (1995): Adenosine, adenosine receptors andthe actions of caffeine. Pharmacol Toxicol 76:93Ð101Fuxe K, Agnati L (1987): Receptor-receptor interactions.Wenner-Gren Center International Symposium Series.Southampton, MacMillan Press.Fuxe K, Agnati LF, Benfenati F, Celani M, Zini I, Zoli M,Mutt V (1983): Evidence for the existence of receptorÐreceptor interactions in the central nervous system.Studies on the regulation of monoamine receptors byneuropeptides. J Neur Transm 18:165Ð179GinŽs S, Hillion J, Torvinen M, Le Crom S, Casado V, CanelaEI, Rondin S, Lew JY, Watson S, Zoli M, Agnati L, VernierP, Lluis C, FerrŽ S, Fuxe K, Franco R (2000): Dopamine D1and adenosine A1 receptors assemble into functionallyinteracting heteromeric complexes. Proc Natl Acad SciUSA 97:8606Ð8611Gouldson PR, Snell CR, Reynolds CA (1997): A new S58R. Franco et al.NEUROPSYCHOPHARMACOLOGY 2000ÐVOL. 23, NO. S4 approach to docking in the beta 2-adrenergic receptorthat exploits the domain structure of G-protein-coupledreceptors. J Med Chem 40:3871Ð3886Gouldson PR, Snell CR, Bywater RP, Higgs C, Reynolds CA(1998): Domain swapping in G-protein coupled receptordimers. Protein Engineering 11:1181Ð1193Hebert TE, Moffett S, Morello JP, Loise TP, Bichet DG, BarretC, Bouvier M (1996): A peptide derived from a beta2-adrenergic receptor transmembrane domain inhibitsboth receptor dimerization and activation. J Biol Chem271:16384Ð16392Jones KA, Borowsky B, Tamm JA, Craig DA, Durkin MM,Dai M, Yao WJ, Johnson M, Gunwaldsen C, Huang LY,Tang C, Shen Q, Salon JA, Morse K, Laz T, Smith KE,Nagarathnam D, Noble SA, Branchek TA, Gerald C(1999): GABAB receptors function as a heteromericassembly of the subunits GABAB R1 and GABAB R2.Nature 396:674Ð679Jordan BA, Devi LA (1999): G protein-coupled receptor het-erodimerization modulates receptor function. Nature399:697Ð700.Kaupmann K, Malitschek B, Schuler V, Heid J, Froestl W,Beck P, Mosbacher J, Bischoff S, Kulik A, Shigemoto R,Karschin A, Bettler B (1999): GABAB-receptor subtypesassemble into functional heteromeric complexes.Nature 396:683Ð687Kull B, FerrŽ S, Arslan G, Svenningsson P, Fuxe K, OwmanC, Fredholm BB (1999): Reciprocal interactions betweenadenosine A2A and dopamine D2 receptors in CHO cellsco-transfected with the two receptors. Biochem Pharma-col 58:1035Ð1045Liu F, Wang Q, Pristupa ZB, Yu XM, Wang YT, Niznik HB(2000): Direct protein-protein coupling enables cross-talk between dopamine D5 and gamma-aminobutyricacid A receptors Nature 403:274Ð280Lohse MJ, Lenschow V, Swabe U (1984): Two affinity statesof Ri adenosine receptors in brain membranes. MolPharmacol 26:1Ð9Maggio R, Vogel Z, Wess J (1993): Coexpression studies withmutant muscarinic/adrenergic receptors provide evi-dence for intermolecular Òcross-talkÓ between G-protein-linked receptors. Proc Natl Acad Sci USA 90:3103Ð3107Maggio R, Barbier P, Fornai F, Corsini GU (1996): Functionalrole of the third cytoplasmatic loop in muscarinic recep-tor dimerization. J Biol Chem 271:31055Ð31060Ng GY, George SR, Zastawny RL, Caron M, Bouvier M,Dennis M, OÕDowd BF (1993): Human serotonin1Breceptor expression in Sf9 cells: phosphorylation, palm-itoylation, and adenylyl cyclase inhibition. Biochemis-try 32:11727Ð11733Ng GY, OÕDowd BF, Caron M, Dennis M, Brann MR, GeorgeSR (1994a): Phosphorylation and palmitoylation of thehuman D2L dopamine receptor in Sf9 cells. J Neuro-chem 63:1589Ð1595Ng GY, Mouillac B, George SR, Caron M, Dennis M, BouvierM, OÕDowd BF (1994b): Desensitization, phosphoryla-tion and palmitoylation of the human dopamine D1receptor. Eur J Pharmacol 267:7Ð19Ng GY, OÕDowd BF, Lee SP, Chung HT, Brann S, Seeman P,George SR (1996): Dopamine D2 receptor dimers andreceptor-blocking peptides. Biochim Biophys Res Com-mun 227:200Ð204Pauwels PJ, Dupuis DS

10 , Perez M, Halazy S (1998): Dimeriza-tio
, Perez M, Halazy S (1998): Dimeriza-tion of 8-OH-DPAT increases activity at serotonin 5-HT1A receptors. Naunyn SchmiedebergÕs Arch Phar-macol 358:404Ð410Phillis JW, Wu PH (1981): The role of adenosine and itsnucleotides in central synpatic transmission. Prog Neu-robiol 16:187Ð239Rivkees SA, Price SL, Zhou FC (1995): Immunohistochemi-cal detection of A1 adenosine receptors in rat brain withemphasis in localization in the hippocampal formation,cerebral cortex, cerebellum, and basal ganglia. Brain Res677:193Ð203Rodriguez-Frade JM, Vila-Coro AJ, Martin A, Albar JP, Mar-t’nez C, Mellado M (1999): The chemokine monocytechemoattractant protein-1 induces functional responsesthrough dimerization of its receptor CCR2. Proc NatlAcad Sci 96:3628Ð3633Romano C, Yang WL, OÕMalley KL (1996): Metabotropicglutamate receptor 5 is a disulfide-linked dimer. J BiolChem 271:28612Ð28616Salim H, FerrŽ S, Dalal A, Peterfreund RA, Fuxe K, VincentD, Lled— PM (2000): Activation of adenosine A1 andA2A receptors modulates dopamine D2 receptor-induced responses in stably transfected human neuro-blastoma cells. J Neurochem 74:432Ð4349Sarri— S, Casad— V, Escriche M, Ciruela, F, Mallol J, Canela,EI, Lluis C, Franco R (2000): The heat shock cognate pro-tein hsc73 assembles with A1 adenosine receptors toform functional modules in the cell membrane. Mol CellBiol 20:5164Ð5174Schiffmann SN, Jacobs O, Vanderhaeghen JJ (1991): Striatalrestricted adenosine A2 receptor (RDC8) is expressedby enkephalin but not by substance P neurons: an insitu hybridization histochemistry study. J Neurochem57:1062Ð1067Schlessinger J (1988): Signal transduction by allosteric recep-tor oligomerization. Trends Biochem Sci 13:443Ð447Schlessinger J, Ullrich A. (1992) Growth factor signaling byreceptor tyrosine kinase. Neuron 9:383Ð391Ullrich A, Schlessinger J (1990): Signal transduction byreceptors with tyrosine kinase activity. Cell 61:203Ð212Venter JC, Fraser CM (1983): beta-Adrenergic receptor isola-tion and characterization with immobilized drugs andmonoclonal antibodies. Fed Proc 42:273Ð278White JH, Wise A, Main MJ, Green A, Fraser NJ, Disney GH,Barnes AA, Emson P, Foord SM, Marshall FH (1999):Heterodimerization is required for the formation of afunctional GABA B receptor. Nature 396:679Ð682Williams M, Risley E (1980a): Biochemical characterizationof putative central purinergic receptors using [3H]2-chloroadenosine, a stable analog of adenosine. ProcNatl Acad Sci USA 77:6892Ð6896Williams M, Risley E (1980b): High affinity binding of 2-chloroadenosine to rat brain synaptic membranes. Eur JPharmacol 64:369Ð370Wilson DK, Rudolph FB, Quiocho FA (1991): Atomic struc-ture of adenosine deaminase complexed with a transition-state analog: understanding catalysis and immunodefi-cience mutations. Science 252:1278Ð1284Wu PH, Phillis JW, Balls K, Rinaldi B. (1980): Specific bind- NEUROPSYCHOPHARMACOLOGY 2000ÐVOL. 23, NO. S4Adenosine/Dopamine Receptor InteractionsS59 ing of [3H]2-chloroadenosine to rat brain cortical mem-branes. Can J Physiol Pharmacol 58: 575Ð579Wu PH, Phillis JW (1982): Adenosine receptors in rat brainmembranes: characterization of high affinity binding of[3H]2-chloroadenosine. Int J Biochem 14:399Ð402Xie Z, Lee SP, OÕDowd BF, George SR (1999): Serotonin 5-HT 1B and 5-HT 1D receptors form homodimers whenexpressed alone and heterodimers when coexpressed.FEBS Lett 456:63Ð67Zawarynski P, Tallerico T, Seeman P, Lee SP, OÕDowd BE,George SR (1998): Dopamine D2 receptor dimers inhuman and rat brain. FEBS Lett 441:383Ð386Zeng FY, Wess J (1999): Identification and molecular charac-terization of m3 muscarininc receptor dimers. J BiolChem 274:19487Ð19497Zoli M, Agnati LF, Hedlund PB, Li XM, FerrŽ S, Fuxe K(1993): Receptor-receptor interactions as an integrativemechanism in nerve cells. Mol Neurobiol 7:29