- Email: [email protected]
Cannabinoid receptor homo- and heterodimerization
Life Sciences 77 (2005) 1667 – 1673 www.elsevier.com/locate/lifescie
Cannabinoid receptor homo- and heterodimerization Ken MackieT Departments of Anesthesiology and Physiology and Biophysics, Box 356540, University of Washington, Seattle, WA 98195-6540, USA
Abstract CB1 cannabinoid receptors mediate the psychoactive effects of D9THC and actions of the endogenous cannabinoids [Howlett, A.C., Barth, F., Bonner, T.I., Cabral, G., Casellas, P., Devane, W.A., Felder, C.C., Herkenham, M., Mackie, K., Martin, B.R., Mechoulam, R., Pertwee, R.G., 2002. International Union of Pharmacology: XXVII. Classification of cannabinoid receptors. Pharmacological Reviews 54 (2) 161–202.]. CB1 receptors belong to the G protein-coupled receptor (GPCR) superfamily. In recent years, it has become apparent that many GPCRs exist as multimers—either of like or unlike receptors [Kroeger, K.M., Pfleger, K.D., Eidne, K.A., 2003. G-protein coupled receptor oligomerization in neuroendocrine pathways. Frontiers of Neuroendocrinology 24 (4) 254–278; Milligan, G., 2004. G protein-coupled receptor dimerization: function and ligand pharmacology. Molecular Pharmacology 66 (1) 1–7.]. Importantly, GPCR multimerization plays a key role in enriching the signaling repertoire of these receptors. In this review, the evidence for CB1 multimerization will be presented, the implications for cannabinoid signaling discussed, and possible future directions for this research considered. D 2005 Elsevier Inc. All rights reserved. Keywords: CB1 receptor; G protein-coupled receptor (GPCR); Multimerization
Introduction Classically, G protein-coupled receptors (GPCRs) were thought to be expressed as monomers and that each individual GPCR signaled through its own pool of G proteins. Recent work from a number of investigators clearly shows this is not the case. (Review of early studies investigating binding cooperativity and radiation inactivation also support the concept of functional GPCRs as multimers.) T Tel.: +1 206 616 2669; fax: +1 206 543 2958. E-mail address: [email protected]. 0024-3205/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2005.05.011
1668
K. Mackie / Life Sciences 77 (2005) 1667–1673
These studies have shown that GPCRs can exist and function as dimers or higher order multimers. (Throughout this review, dimerization and multimerization are used interchangeably in deference to our incomplete knowledge of the tertiary structure of the receptor complexes.) Most dimerization studies have used immunoprecipitation (Jordan and Devi, 1999), complementation (Scarselli et al., 2001), or resonance energy transfer (Canals et al., 2004) approaches to demonstrate association. Typically, for the immunoprecipitation studies, two receptors with distinct epitope tags are heterologously expressed in cell lines. The cells are then lysed and one receptor is immunoprecipitated with antibodies directed against its first epitope tag. The immunoprecipitated sample is then subjected to Western blotting with an antibody directed against the second tag. If the appropriate controls are conducted, and the second antibody recognizes the tagged receptor in the immunoprecipitate, then this is taken for evidence of association of the two receptors. In complementation studies, two receptors are expressed (either with point mutations or truncations) that are by themselves inactive. If co-expression of the two receptors together reconstitutes a functional receptor, this is taken as evidence that the two proteins are interacting. An interesting feature of these complementation studies is that they have identified dominant negative or dominant positive interactions between pairs of GPCRs. The best known example of this is the interaction between CCR5 and its naturally occurring truncation CCR5D32 (Kazmierski et al., 2002). Expression of the truncated CCR5 dramatically slows wild-type CCR5 internalization (necessary for infection by M-tropic HIV). The resonance energy transfer approaches, typically FRET (fluorescent resonance energy transfer) or BRET (bioluminescence resonance energy transfer), rely on the observation that non-radiative transfer of light between donor and acceptor molecules is proportional to the reciprocal of the separating distance raised to the sixth power (Jares-Erijman and Jovin, 2003). Thus, receptors whose donor and acceptors are ˚ ) will undergo resonance transfer, while more distant receptors will not. In close to one another (b50–100 A FRET, the donor and acceptor molecules are coupled to the receptors (for example, by the incorporation of YFP and CFP in the primary structure of the receptors). For BRET, the enzyme luciferase is incorporated into one receptor (which will catalyze the conversion of coelenterazine to coelenteramide with the generation of light with a peak wavelength of ~ 470 nm) (Pfleger and Eidne, 2003). YFP in the second receptor excited by the light generated during the degradation of coelenterazine will then fluoresce with a peak of ~ 530 nm. An advantage of BRET is that no excitation of the donor is necessary, thus avoiding the problems associated with exposing living cells to relatively high-intensity, low-wavelength light such as autofluorescence, photobleaching, and phototoxicity. A drawback of BRET is that with current instrumentation and techniques, spatial resolution is poor. In contrast, with optimal conditions, FRET can be detected in subcellular domains. For the resonance energy transfer techniques, it is important to consider that not only the distance, but also the relative orientations of the donor and acceptor are important, particularly for FRET experiments. Thus, decreases in FRET in GPRC dimerization studies could be due to either an increase in the distance separating the receptors or due to the assumption of an unfavorable conformation between donor and the acceptor, in the absence of a change in the absolute separation. Thus, the most robust results will come from studies that use independent methods to determine receptor association. By now, a great number of studies have been conducted using these techniques investigating many GPCRs. While there is some discrepancy between various studies, several common themes have emerged: (1) GPCR homodimerization is very common; likely all GPCRs will be able to form homodimers.
K. Mackie / Life Sciences 77 (2005) 1667–1673
1669
(2) GPCR heterodimerization also occurs, both between receptors that are members of the same class as well as between members of different classes. (3) GPCR heterodimerization can lead to binding sites that bind ligands recognized by neither receptor when expressed individually. (4) GPCR heterodimerization can give rise to receptor complexes that signal and/or traffic differently from the component GPCRs. (5) There is no single bdimerization domain.Q Dimerization occurs via extracellular domains, transmembrane helices, and intracellular domains, depending on the receptor. (6) GPCRs first associate during receptor synthesis. Furthermore, the degree and regulation of dimerization by agonists at the cell surface is quite variable among receptors. (7) GPCRs can also associate with ionotropic receptors, leading to alterations in the function of both receptors.
CB1 receptor homodimerization CB1 receptors form homodimers. This has been demonstrated using an antibody that preferentially recognizes the dimerized form of the receptor (Fig. 1) as well as by the more classical coimmunoprecipitation technique (Wager-Miller et al., 2002). Using this bdimer antibody,Q immunohistochemical studies demonstrated that CB1 receptors exist as dimers in the brain (Hajos et al., 2000; Katona et al., 2001). Interestingly, the staining pattern seen with this antibody is grossly similar to the pattern seen with CB1 antibodies that do not distinguish between the monomeric and multimeric forms of the receptor. One interpretation of this observation is that there is not a preferential localization of
Fig. 1. High-molecular-weight forms of CB1 exist in situ. (A) Western blot of rat cortical membranes probed with an amino terminal CB1 antibody (AT) and a carboxy-terminal CB1 antibody (CT). The amino terminal antibody recognizes both a highand low-molecular-weight form of CB1 (arrowheads), while the carboxy-terminal antibody recognizes only the high-molecularweight form. AT staining of CB1 is blocked by the immunizing protein (GDT-AT) and CT staining is blocked by the immunizing protein (GST-CT). (B) Immunostaining of rat dentate gyrus with carboxy terminus antibody reveals the same axonal staining pattern as is seen with amino terminal antibodies. Abbreviations: gc= granule cell layer, oml= outer molecular layer, iml = inner molecular layer. Scale bar = 50 Am.
1670
K. Mackie / Life Sciences 77 (2005) 1667–1673
monomeric CB1 receptors (Katona et al., 1999). Another interpretation consistent with these data is that all CB1 receptors exist as dimers or higher order multimers. The anatomical studies also emphasize the point that GPCRs exist as dimers in brain tissue and decreases the (remote) possibility that dimerization is an artifact of isolation (in the immunoprecipitation experiments) or overexpression (the resonance energy and immunoprecipitation experiments). CB1 association has also been demonstrated by immunoprecipitation experiments. These experiments were done in the classical way—co-expression of differentially epitope-tagged CB1 receptors in cell lines, followed by immunoprecipitation and Western blotting. In addition to confirming that CB1 receptors form homodimers, these experiments also suggest that agonists can regulate the extent of CB1 dimerization (J. Wager-Miller, unpublished observations). These latter results are consistent with those seen in cells transiently transfected with CB1 receptors and treated with cannabinoid agonists using the bdimer antibodyQ (K. Mackie, unpublished results).
CB1 as heterodimers CB1 and D2 dopamine receptor heterodimers CB1 receptors also form heterodimers. The best described CB1 heterodimers are the heterodimers between D2 and CB1 receptors (Kearn et al., 2004). For these receptors, heterodimerization was demonstrated by sequential immunoprecipitation and Western blotting. While heterodimerization was seen in the absence of agonists, CB1 agonists tended to increase, while a CB1 inverse agonist tended to decrease association of the two receptors. While physical association of GPCRs has been widely demonstrated, of more importance are the functional consequences of the association. This is what makes the association between D2 and CB1 receptors particularly interesting. Glass and Felder demonstrated both in transfected cell lines and cultured striatal neurons that while, individually, both D2 and CB1 agonists decreased forskolinstimulated adenylyl cyclase activity, when they were applied together, increasing concentrations of CB1 agonist reversed the adenylyl cyclase inhibition produced by D2 agonists (Glass and Felder, 1997). Furthermore, recent results from my lab suggest a similar interaction is seen between CB1 and D2 receptors in MAP kinase activation (Kearn et al., 2004). In these experiments, co-expression of CB1 and D2 receptors dramatically increased MAP kinase activation by both CB1 and D2 agonists. Interestingly, this increase was independent of PTX-sensitive G proteins (unlike MAP kinase stimulation in cells transfected with a single receptor) suggesting the recruitment of PTX insensitive G proteins by the receptor dimer. Furthermore, the EC50s for these multiple effects (dimer increase, stimulation of adenylyl cyclase, and activation of MAP kinase) are all similar, consistent with a related mechanism of action such as the involvement of dimers in the unique signaling characteristics of the cells expressing CB1 and D2 receptors. CB1 and opioid receptor heterodimers CB1 and opioid receptors are often co-expressed in the same subcellular compartments (Pickel et al., 2004; Rodriguez et al., 2001; Salio et al., 2001). Thus, it is interesting that CB1 receptors can also form heterodimers with several of the opioid receptors (Rios et al., 2002). The functional implications of
K. Mackie / Life Sciences 77 (2005) 1667–1673
1671
opioid-CB1 receptor heterodimerization are not certain; however, altered activation of MAP kinase by cannabinoid and opioid receptor agonists has been described (Rios et al., 2002). Implications of CB1 receptor dimerization While CB1 receptors can clearly form homo- and heterodimers, many important questions remain. In the case of CB1 homodimers, unresolved questions include the following. What are the dimerization domains for the CB1 receptor? What are the implications of agonist-induced dissociation of CB1 dimers? Once dissociated, do CB1 dimers reform? If so, in what cellular compartment re-associate? How do CB1 dimers signal? Does the dimer form a signaling unit; in particular, how does dimer formation affect the ligand binding site (almost uniformly modeled in the monomer conformation), and what receptor surface is presented to the G protein interacting with the receptor (evidence from rhodopsin suggests a single G protein might interact with the dimer)? Similarly, there are a number of interesting unresolved issues with CB1 heterodimers. In addition to the questions above that are relevant for heterodimers, there are also some that are specific for heterodimers. Perhaps the most interesting ones are which other GPCRs do CB1 receptors dimerize with and what the functional consequences are. One exciting possibility is that dimerization of CB1 receptors with other GPCRs accounts for some of the bnon-traditionalQ CB1 receptor signaling reported in the literature. An example of this would be the well-characterized mobilization of intracellular calcium by a wide range of CB1 agonists in the NG108-15 neuroblastoma glioma cell line (Sugiura et al., 1996, 1999). It is possible that this phenomenon requires the dimerization of CB1 with another, possibly Gq/11-linked GPCR. In addition, some of the vascular response to cannabinoids, while mediated by CB1 receptors, have unique attributes that could be explained by the heterodimerization of CB1 and other GPCRs (Kunos et al., 2002; Randall et al., 2002). An interesting, and not yet studied, possibility is that CB1 receptors heterodimerize with other receptors for endogenous lipids, such as the EDG or sphingosphine 1-phosphate receptors (Kostenis, 2004). Such heterodimerization may give rise to lipid receptors with unique ligand-binding properties and signaling characteristics. Finally, a particularly intriguing class of CB1 heterodimerization would be with ionotropic receptors. Possible examples (based on partial anatomical or functional co-localization) would be with TRPV1, glutamate NR2B, and serotonin 5-HT3 receptors (Bridges et al., 2003; Hermann et al., 2002; Sjostrom et al., 2003). The importance of dimerization in GPCR signaling is becoming well established. Compelling evidence suggests that CB1 receptors form both homo- and heterodimers. However, much work remains to be done to determine to what extent these dimers contribute to the complexity of cannabinoid signaling. Undoubtedly, in the next few years, exciting results will be forthcoming that will cause us to revise our linear view of cannabinoid receptor signaling. References Bridges, D., Rice, A.S., Egertova, M., Elphick, M.R., Winter, J., Michael, G.J., 2003. Localisation of cannabinoid receptor 1 in rat dorsal root ganglion using in situ hybridisation and immunohistochemistry. Neuroscience 119 (3), 803 – 812. Canals, M., Burgueno, J., Marcellino, D., Cabello, N., Canela, E.I., Mallol, J., Agnati, L., Ferre, S., Bouvier, M., Fuxe, K., Ciruela, F., Lluis, C., Franco, R., 2004. Homodimerization of adenosine A2A receptors: qualitative and quantitative assessment by fluorescence and bioluminescence energy transfer. Journal of Neurochemistry 88 (3), 726 – 734.
1672
K. Mackie / Life Sciences 77 (2005) 1667–1673
Glass, M., Felder, C.C., 1997. Concurrent stimulation of cannabinoid CB1 and dopamine D2 receptors augments cAMP accumulation in striatal neurons: evidence for a Gs linkage to the CB1 receptor. Journal of Neuroscience 17 (14), 5327 – 5333. Hajos, N., Katona, I., Naiem, S.S., Mackie, K., Ledent, C., Mody, I., Freund, T.F., 2000. Cannabinoids inhibit hippocampal GABAergic transmission and network oscillations. European Journal of Neuroscience 12 (9), 3239 – 3249. Hermann, H., Marsicano, G., Lutz, B., 2002. Coexpression of the cannabinoid receptor type 1 with dopamine and serotonin receptors in distinct neuronal subpopulations of the adult mouse forebrain. Neuroscience 109 (3), 451 – 460. Howlett, A.C., Barth, F., Bonner, T.I., Cabral, G., Casellas, P., Devane, W.A., Felder, C.C., Herkenham, M., Mackie, K., Martin, B.R., Mechoulam, R., Pertwee, R.G., 2002. International union of pharmacology: XXVII. Classification of cannabinoid receptors. Pharmacological Reviews 54 (2), 161 – 202. Jares-Erijman, E.A., Jovin, T.M., 2003. FRET imaging. Nature Biotechnology 21 (11), 1387 – 1395. Jordan, B.A., Devi, L.A., 1999. G-protein-coupled receptor heterodimerization modulates receptor function. Nature 399 (6737), 697 – 700. Katona, I., Sperlagh, B., Sik, A., Kafalvi, A., Vizi, E.S., Mackie, K., Freund, T.F., 1999. Presynaptically located CB1 cannabinoid receptors regulate GABA release from axon terminals of specific hippocampal interneurons. Journal of Neuroscience 19 (11), 4544 – 4558. Katona, I., Rancz, E.A., Acsady, L., Ledent, C., Mackie, K., Hajos, N., Freund, T.F., 2001. Distribution of CB1 cannabinoid receptors in the amygdala and their role in the control of GABAergic transmission. Journal of Neuroscience 21 (23), 9506 – 9518. Kazmierski, W.M., Boone, L., Lawrence, W., Watson, C., Kenakin, T., 2002. CCR5 chemokine receptors: Gatekeepers of HIV1 infection. Current Drug Targets of Infectious Disorders 2 (3), 265 – 278. Kearn, C.S., Mackie, K., Glass, M., 2004. Physical interactions of CB1 cannabinoid and D2 receptors. 2004 Symposium of the Cannabinoids, p. 16. Kostenis, E., 2004. Novel clusters of receptors for sphingosine-1-phosphate, sphingosylphosphorylcholine, and (lyso)phosphatidic acid: new receptors for boldQ ligands. Journal of Cellular Biochemistry 92 (5), 923 – 936. Kroeger, K.M., Pfleger, K.D., Eidne, K.A., 2003. G-protein coupled receptor oligomerization in neuroendocrine pathways. Frontiers of Neuroendocrinology 24 (4), 254 – 278. Kunos, G., Batkai, S., Offertaler, L., Mo, F., Liu, J., Karcher, J., Harvey-White, J., 2002. The quest for a vascular endothelial cannabinoid receptor. Chemistry and Physics of Lipids 121 (1–2), 45 – 56. Milligan, G., 2004. G protein-coupled receptor dimerization: function and ligand pharmacology. Molecular Pharmacology 66 (1), 1 – 7. Pfleger, K.D., Eidne, K.A., 2003. New technologies: bioluminescence resonance energy transfer (BRET) for the detection of real time interactions involving G-protein coupled receptors. Pituitary 6 (3), 141 – 151. Pickel, V.M., Chan, J., Kash, T.L., Rodriguez, J.J., Mackie, K., 2004. Compartment-specific localization of cannabinoid 1 (CB1) and mu-opioid receptors in rat nucleus accumbens. Neuroscience 127 (1), 101 – 112. Randall, M.D., Harris, D., Kendall, D.A., Ralevic, V., 2002. Cardiovascular effects of cannabinoids. Pharmacology and Therapeutics 95 (2), 191 – 202. Rios, C., Haberny, S., Gomes, I., Trapaidze, N., Devi, L.A., 2002. Opioid-cannabinoid cross talk: a role for receptor–receptor interactions. 2002 Symposium on the Cannabinoids, p. 21. Rodriguez, J.J., Mackie, K., Pickel, V.M., 2001. Ultrastructural localization of the CB1 cannabinoid receptor in mu-opioid receptor patches of the rat caudate putamen nucleus. Journal of Neuroscience 21 (3), 823 – 833. Salio, C., Fischer, J., Franzoni, M.F., Mackie, K., Kaneko, T., Conrath, M., 2001. CB1-cannabinoid and mu-opioid receptor colocalization on postsynaptic target in the rat dorsal horn. NeuroReport 12 (17), 3689 – 3692. Scarselli, M., Novi, F., Schallmach, E., Lin, R., Baragli, A., Colzi, A., Griffon, N., Corsini, G.U., Sokoloff, P., Levenson, R., Vogel, Z., Maggio, R., 2001. D2/D3 dopamine receptor heterodimers exhibit unique functional properties. Journal of Biological Chemistry 276 (32), 30308 – 30314. Sjostrom, P.J., Turrigiano, G.G., Nelson, S.B., 2003. Neocortical LTD via coincident activation of presynaptic NMDA and cannabinoid receptors. Neuron 39 (4), 641 – 654. Sugiura, T., Kodaka, T., Kondo, S., Tonegawa, T., Nakane, S., Kishimoto, S., Yamashita, A., Waku, K., 1996. 2Arachidonoylglycerol, a putative endogenous cannabinoid receptor ligand, induces rapid, transient elevation of intracellular free Ca2+ in neuroblastoma glioma hybrid NG108-15 cells. Biochemistry and Biophysical Research Communications 229 (1), 58 – 64.
K. Mackie / Life Sciences 77 (2005) 1667–1673
1673
Sugiura, T., Kodaka, T., Nakane, S., Miyashita, T., Kondo, S., Suhara, Y., Takayama, H., Waku, K., Seki, C., Baba, N., Ishima, Y., 1999. Evidence that the cannabinoid CB1 receptor is a 2-arachidonoylglycerol receptor. Structure-activity relationship of 2-arachidonoylglycerol, ether-linked analogues, and related compounds. Journal of Biological Chemistry 274 (5), 2794 – 2801. Wager-Miller, J., Westenbroek, R., Mackie, K., 2002. Dimerization of G protein-coupled receptors: CB1 cannabinoid receptors as an example. Chemistry and Physics of Lipids 121 (1–2), 83 – 89.
Cannabinoid receptor homo- and heterodimerization Ken MackieT Departments of Anesthesiology and Physiology and Biophysics, Box 356540, University of Washington, Seattle, WA 98195-6540, USA
Abstract CB1 cannabinoid receptors mediate the psychoactive effects of D9THC and actions of the endogenous cannabinoids [Howlett, A.C., Barth, F., Bonner, T.I., Cabral, G., Casellas, P., Devane, W.A., Felder, C.C., Herkenham, M., Mackie, K., Martin, B.R., Mechoulam, R., Pertwee, R.G., 2002. International Union of Pharmacology: XXVII. Classification of cannabinoid receptors. Pharmacological Reviews 54 (2) 161–202.]. CB1 receptors belong to the G protein-coupled receptor (GPCR) superfamily. In recent years, it has become apparent that many GPCRs exist as multimers—either of like or unlike receptors [Kroeger, K.M., Pfleger, K.D., Eidne, K.A., 2003. G-protein coupled receptor oligomerization in neuroendocrine pathways. Frontiers of Neuroendocrinology 24 (4) 254–278; Milligan, G., 2004. G protein-coupled receptor dimerization: function and ligand pharmacology. Molecular Pharmacology 66 (1) 1–7.]. Importantly, GPCR multimerization plays a key role in enriching the signaling repertoire of these receptors. In this review, the evidence for CB1 multimerization will be presented, the implications for cannabinoid signaling discussed, and possible future directions for this research considered. D 2005 Elsevier Inc. All rights reserved. Keywords: CB1 receptor; G protein-coupled receptor (GPCR); Multimerization
Introduction Classically, G protein-coupled receptors (GPCRs) were thought to be expressed as monomers and that each individual GPCR signaled through its own pool of G proteins. Recent work from a number of investigators clearly shows this is not the case. (Review of early studies investigating binding cooperativity and radiation inactivation also support the concept of functional GPCRs as multimers.) T Tel.: +1 206 616 2669; fax: +1 206 543 2958. E-mail address: [email protected]. 0024-3205/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2005.05.011
1668
K. Mackie / Life Sciences 77 (2005) 1667–1673
These studies have shown that GPCRs can exist and function as dimers or higher order multimers. (Throughout this review, dimerization and multimerization are used interchangeably in deference to our incomplete knowledge of the tertiary structure of the receptor complexes.) Most dimerization studies have used immunoprecipitation (Jordan and Devi, 1999), complementation (Scarselli et al., 2001), or resonance energy transfer (Canals et al., 2004) approaches to demonstrate association. Typically, for the immunoprecipitation studies, two receptors with distinct epitope tags are heterologously expressed in cell lines. The cells are then lysed and one receptor is immunoprecipitated with antibodies directed against its first epitope tag. The immunoprecipitated sample is then subjected to Western blotting with an antibody directed against the second tag. If the appropriate controls are conducted, and the second antibody recognizes the tagged receptor in the immunoprecipitate, then this is taken for evidence of association of the two receptors. In complementation studies, two receptors are expressed (either with point mutations or truncations) that are by themselves inactive. If co-expression of the two receptors together reconstitutes a functional receptor, this is taken as evidence that the two proteins are interacting. An interesting feature of these complementation studies is that they have identified dominant negative or dominant positive interactions between pairs of GPCRs. The best known example of this is the interaction between CCR5 and its naturally occurring truncation CCR5D32 (Kazmierski et al., 2002). Expression of the truncated CCR5 dramatically slows wild-type CCR5 internalization (necessary for infection by M-tropic HIV). The resonance energy transfer approaches, typically FRET (fluorescent resonance energy transfer) or BRET (bioluminescence resonance energy transfer), rely on the observation that non-radiative transfer of light between donor and acceptor molecules is proportional to the reciprocal of the separating distance raised to the sixth power (Jares-Erijman and Jovin, 2003). Thus, receptors whose donor and acceptors are ˚ ) will undergo resonance transfer, while more distant receptors will not. In close to one another (b50–100 A FRET, the donor and acceptor molecules are coupled to the receptors (for example, by the incorporation of YFP and CFP in the primary structure of the receptors). For BRET, the enzyme luciferase is incorporated into one receptor (which will catalyze the conversion of coelenterazine to coelenteramide with the generation of light with a peak wavelength of ~ 470 nm) (Pfleger and Eidne, 2003). YFP in the second receptor excited by the light generated during the degradation of coelenterazine will then fluoresce with a peak of ~ 530 nm. An advantage of BRET is that no excitation of the donor is necessary, thus avoiding the problems associated with exposing living cells to relatively high-intensity, low-wavelength light such as autofluorescence, photobleaching, and phototoxicity. A drawback of BRET is that with current instrumentation and techniques, spatial resolution is poor. In contrast, with optimal conditions, FRET can be detected in subcellular domains. For the resonance energy transfer techniques, it is important to consider that not only the distance, but also the relative orientations of the donor and acceptor are important, particularly for FRET experiments. Thus, decreases in FRET in GPRC dimerization studies could be due to either an increase in the distance separating the receptors or due to the assumption of an unfavorable conformation between donor and the acceptor, in the absence of a change in the absolute separation. Thus, the most robust results will come from studies that use independent methods to determine receptor association. By now, a great number of studies have been conducted using these techniques investigating many GPCRs. While there is some discrepancy between various studies, several common themes have emerged: (1) GPCR homodimerization is very common; likely all GPCRs will be able to form homodimers.
K. Mackie / Life Sciences 77 (2005) 1667–1673
1669
(2) GPCR heterodimerization also occurs, both between receptors that are members of the same class as well as between members of different classes. (3) GPCR heterodimerization can lead to binding sites that bind ligands recognized by neither receptor when expressed individually. (4) GPCR heterodimerization can give rise to receptor complexes that signal and/or traffic differently from the component GPCRs. (5) There is no single bdimerization domain.Q Dimerization occurs via extracellular domains, transmembrane helices, and intracellular domains, depending on the receptor. (6) GPCRs first associate during receptor synthesis. Furthermore, the degree and regulation of dimerization by agonists at the cell surface is quite variable among receptors. (7) GPCRs can also associate with ionotropic receptors, leading to alterations in the function of both receptors.
CB1 receptor homodimerization CB1 receptors form homodimers. This has been demonstrated using an antibody that preferentially recognizes the dimerized form of the receptor (Fig. 1) as well as by the more classical coimmunoprecipitation technique (Wager-Miller et al., 2002). Using this bdimer antibody,Q immunohistochemical studies demonstrated that CB1 receptors exist as dimers in the brain (Hajos et al., 2000; Katona et al., 2001). Interestingly, the staining pattern seen with this antibody is grossly similar to the pattern seen with CB1 antibodies that do not distinguish between the monomeric and multimeric forms of the receptor. One interpretation of this observation is that there is not a preferential localization of
Fig. 1. High-molecular-weight forms of CB1 exist in situ. (A) Western blot of rat cortical membranes probed with an amino terminal CB1 antibody (AT) and a carboxy-terminal CB1 antibody (CT). The amino terminal antibody recognizes both a highand low-molecular-weight form of CB1 (arrowheads), while the carboxy-terminal antibody recognizes only the high-molecularweight form. AT staining of CB1 is blocked by the immunizing protein (GDT-AT) and CT staining is blocked by the immunizing protein (GST-CT). (B) Immunostaining of rat dentate gyrus with carboxy terminus antibody reveals the same axonal staining pattern as is seen with amino terminal antibodies. Abbreviations: gc= granule cell layer, oml= outer molecular layer, iml = inner molecular layer. Scale bar = 50 Am.
1670
K. Mackie / Life Sciences 77 (2005) 1667–1673
monomeric CB1 receptors (Katona et al., 1999). Another interpretation consistent with these data is that all CB1 receptors exist as dimers or higher order multimers. The anatomical studies also emphasize the point that GPCRs exist as dimers in brain tissue and decreases the (remote) possibility that dimerization is an artifact of isolation (in the immunoprecipitation experiments) or overexpression (the resonance energy and immunoprecipitation experiments). CB1 association has also been demonstrated by immunoprecipitation experiments. These experiments were done in the classical way—co-expression of differentially epitope-tagged CB1 receptors in cell lines, followed by immunoprecipitation and Western blotting. In addition to confirming that CB1 receptors form homodimers, these experiments also suggest that agonists can regulate the extent of CB1 dimerization (J. Wager-Miller, unpublished observations). These latter results are consistent with those seen in cells transiently transfected with CB1 receptors and treated with cannabinoid agonists using the bdimer antibodyQ (K. Mackie, unpublished results).
CB1 as heterodimers CB1 and D2 dopamine receptor heterodimers CB1 receptors also form heterodimers. The best described CB1 heterodimers are the heterodimers between D2 and CB1 receptors (Kearn et al., 2004). For these receptors, heterodimerization was demonstrated by sequential immunoprecipitation and Western blotting. While heterodimerization was seen in the absence of agonists, CB1 agonists tended to increase, while a CB1 inverse agonist tended to decrease association of the two receptors. While physical association of GPCRs has been widely demonstrated, of more importance are the functional consequences of the association. This is what makes the association between D2 and CB1 receptors particularly interesting. Glass and Felder demonstrated both in transfected cell lines and cultured striatal neurons that while, individually, both D2 and CB1 agonists decreased forskolinstimulated adenylyl cyclase activity, when they were applied together, increasing concentrations of CB1 agonist reversed the adenylyl cyclase inhibition produced by D2 agonists (Glass and Felder, 1997). Furthermore, recent results from my lab suggest a similar interaction is seen between CB1 and D2 receptors in MAP kinase activation (Kearn et al., 2004). In these experiments, co-expression of CB1 and D2 receptors dramatically increased MAP kinase activation by both CB1 and D2 agonists. Interestingly, this increase was independent of PTX-sensitive G proteins (unlike MAP kinase stimulation in cells transfected with a single receptor) suggesting the recruitment of PTX insensitive G proteins by the receptor dimer. Furthermore, the EC50s for these multiple effects (dimer increase, stimulation of adenylyl cyclase, and activation of MAP kinase) are all similar, consistent with a related mechanism of action such as the involvement of dimers in the unique signaling characteristics of the cells expressing CB1 and D2 receptors. CB1 and opioid receptor heterodimers CB1 and opioid receptors are often co-expressed in the same subcellular compartments (Pickel et al., 2004; Rodriguez et al., 2001; Salio et al., 2001). Thus, it is interesting that CB1 receptors can also form heterodimers with several of the opioid receptors (Rios et al., 2002). The functional implications of
K. Mackie / Life Sciences 77 (2005) 1667–1673
1671
opioid-CB1 receptor heterodimerization are not certain; however, altered activation of MAP kinase by cannabinoid and opioid receptor agonists has been described (Rios et al., 2002). Implications of CB1 receptor dimerization While CB1 receptors can clearly form homo- and heterodimers, many important questions remain. In the case of CB1 homodimers, unresolved questions include the following. What are the dimerization domains for the CB1 receptor? What are the implications of agonist-induced dissociation of CB1 dimers? Once dissociated, do CB1 dimers reform? If so, in what cellular compartment re-associate? How do CB1 dimers signal? Does the dimer form a signaling unit; in particular, how does dimer formation affect the ligand binding site (almost uniformly modeled in the monomer conformation), and what receptor surface is presented to the G protein interacting with the receptor (evidence from rhodopsin suggests a single G protein might interact with the dimer)? Similarly, there are a number of interesting unresolved issues with CB1 heterodimers. In addition to the questions above that are relevant for heterodimers, there are also some that are specific for heterodimers. Perhaps the most interesting ones are which other GPCRs do CB1 receptors dimerize with and what the functional consequences are. One exciting possibility is that dimerization of CB1 receptors with other GPCRs accounts for some of the bnon-traditionalQ CB1 receptor signaling reported in the literature. An example of this would be the well-characterized mobilization of intracellular calcium by a wide range of CB1 agonists in the NG108-15 neuroblastoma glioma cell line (Sugiura et al., 1996, 1999). It is possible that this phenomenon requires the dimerization of CB1 with another, possibly Gq/11-linked GPCR. In addition, some of the vascular response to cannabinoids, while mediated by CB1 receptors, have unique attributes that could be explained by the heterodimerization of CB1 and other GPCRs (Kunos et al., 2002; Randall et al., 2002). An interesting, and not yet studied, possibility is that CB1 receptors heterodimerize with other receptors for endogenous lipids, such as the EDG or sphingosphine 1-phosphate receptors (Kostenis, 2004). Such heterodimerization may give rise to lipid receptors with unique ligand-binding properties and signaling characteristics. Finally, a particularly intriguing class of CB1 heterodimerization would be with ionotropic receptors. Possible examples (based on partial anatomical or functional co-localization) would be with TRPV1, glutamate NR2B, and serotonin 5-HT3 receptors (Bridges et al., 2003; Hermann et al., 2002; Sjostrom et al., 2003). The importance of dimerization in GPCR signaling is becoming well established. Compelling evidence suggests that CB1 receptors form both homo- and heterodimers. However, much work remains to be done to determine to what extent these dimers contribute to the complexity of cannabinoid signaling. Undoubtedly, in the next few years, exciting results will be forthcoming that will cause us to revise our linear view of cannabinoid receptor signaling. References Bridges, D., Rice, A.S., Egertova, M., Elphick, M.R., Winter, J., Michael, G.J., 2003. Localisation of cannabinoid receptor 1 in rat dorsal root ganglion using in situ hybridisation and immunohistochemistry. Neuroscience 119 (3), 803 – 812. Canals, M., Burgueno, J., Marcellino, D., Cabello, N., Canela, E.I., Mallol, J., Agnati, L., Ferre, S., Bouvier, M., Fuxe, K., Ciruela, F., Lluis, C., Franco, R., 2004. Homodimerization of adenosine A2A receptors: qualitative and quantitative assessment by fluorescence and bioluminescence energy transfer. Journal of Neurochemistry 88 (3), 726 – 734.
1672
K. Mackie / Life Sciences 77 (2005) 1667–1673
Glass, M., Felder, C.C., 1997. Concurrent stimulation of cannabinoid CB1 and dopamine D2 receptors augments cAMP accumulation in striatal neurons: evidence for a Gs linkage to the CB1 receptor. Journal of Neuroscience 17 (14), 5327 – 5333. Hajos, N., Katona, I., Naiem, S.S., Mackie, K., Ledent, C., Mody, I., Freund, T.F., 2000. Cannabinoids inhibit hippocampal GABAergic transmission and network oscillations. European Journal of Neuroscience 12 (9), 3239 – 3249. Hermann, H., Marsicano, G., Lutz, B., 2002. Coexpression of the cannabinoid receptor type 1 with dopamine and serotonin receptors in distinct neuronal subpopulations of the adult mouse forebrain. Neuroscience 109 (3), 451 – 460. Howlett, A.C., Barth, F., Bonner, T.I., Cabral, G., Casellas, P., Devane, W.A., Felder, C.C., Herkenham, M., Mackie, K., Martin, B.R., Mechoulam, R., Pertwee, R.G., 2002. International union of pharmacology: XXVII. Classification of cannabinoid receptors. Pharmacological Reviews 54 (2), 161 – 202. Jares-Erijman, E.A., Jovin, T.M., 2003. FRET imaging. Nature Biotechnology 21 (11), 1387 – 1395. Jordan, B.A., Devi, L.A., 1999. G-protein-coupled receptor heterodimerization modulates receptor function. Nature 399 (6737), 697 – 700. Katona, I., Sperlagh, B., Sik, A., Kafalvi, A., Vizi, E.S., Mackie, K., Freund, T.F., 1999. Presynaptically located CB1 cannabinoid receptors regulate GABA release from axon terminals of specific hippocampal interneurons. Journal of Neuroscience 19 (11), 4544 – 4558. Katona, I., Rancz, E.A., Acsady, L., Ledent, C., Mackie, K., Hajos, N., Freund, T.F., 2001. Distribution of CB1 cannabinoid receptors in the amygdala and their role in the control of GABAergic transmission. Journal of Neuroscience 21 (23), 9506 – 9518. Kazmierski, W.M., Boone, L., Lawrence, W., Watson, C., Kenakin, T., 2002. CCR5 chemokine receptors: Gatekeepers of HIV1 infection. Current Drug Targets of Infectious Disorders 2 (3), 265 – 278. Kearn, C.S., Mackie, K., Glass, M., 2004. Physical interactions of CB1 cannabinoid and D2 receptors. 2004 Symposium of the Cannabinoids, p. 16. Kostenis, E., 2004. Novel clusters of receptors for sphingosine-1-phosphate, sphingosylphosphorylcholine, and (lyso)phosphatidic acid: new receptors for boldQ ligands. Journal of Cellular Biochemistry 92 (5), 923 – 936. Kroeger, K.M., Pfleger, K.D., Eidne, K.A., 2003. G-protein coupled receptor oligomerization in neuroendocrine pathways. Frontiers of Neuroendocrinology 24 (4), 254 – 278. Kunos, G., Batkai, S., Offertaler, L., Mo, F., Liu, J., Karcher, J., Harvey-White, J., 2002. The quest for a vascular endothelial cannabinoid receptor. Chemistry and Physics of Lipids 121 (1–2), 45 – 56. Milligan, G., 2004. G protein-coupled receptor dimerization: function and ligand pharmacology. Molecular Pharmacology 66 (1), 1 – 7. Pfleger, K.D., Eidne, K.A., 2003. New technologies: bioluminescence resonance energy transfer (BRET) for the detection of real time interactions involving G-protein coupled receptors. Pituitary 6 (3), 141 – 151. Pickel, V.M., Chan, J., Kash, T.L., Rodriguez, J.J., Mackie, K., 2004. Compartment-specific localization of cannabinoid 1 (CB1) and mu-opioid receptors in rat nucleus accumbens. Neuroscience 127 (1), 101 – 112. Randall, M.D., Harris, D., Kendall, D.A., Ralevic, V., 2002. Cardiovascular effects of cannabinoids. Pharmacology and Therapeutics 95 (2), 191 – 202. Rios, C., Haberny, S., Gomes, I., Trapaidze, N., Devi, L.A., 2002. Opioid-cannabinoid cross talk: a role for receptor–receptor interactions. 2002 Symposium on the Cannabinoids, p. 21. Rodriguez, J.J., Mackie, K., Pickel, V.M., 2001. Ultrastructural localization of the CB1 cannabinoid receptor in mu-opioid receptor patches of the rat caudate putamen nucleus. Journal of Neuroscience 21 (3), 823 – 833. Salio, C., Fischer, J., Franzoni, M.F., Mackie, K., Kaneko, T., Conrath, M., 2001. CB1-cannabinoid and mu-opioid receptor colocalization on postsynaptic target in the rat dorsal horn. NeuroReport 12 (17), 3689 – 3692. Scarselli, M., Novi, F., Schallmach, E., Lin, R., Baragli, A., Colzi, A., Griffon, N., Corsini, G.U., Sokoloff, P., Levenson, R., Vogel, Z., Maggio, R., 2001. D2/D3 dopamine receptor heterodimers exhibit unique functional properties. Journal of Biological Chemistry 276 (32), 30308 – 30314. Sjostrom, P.J., Turrigiano, G.G., Nelson, S.B., 2003. Neocortical LTD via coincident activation of presynaptic NMDA and cannabinoid receptors. Neuron 39 (4), 641 – 654. Sugiura, T., Kodaka, T., Kondo, S., Tonegawa, T., Nakane, S., Kishimoto, S., Yamashita, A., Waku, K., 1996. 2Arachidonoylglycerol, a putative endogenous cannabinoid receptor ligand, induces rapid, transient elevation of intracellular free Ca2+ in neuroblastoma glioma hybrid NG108-15 cells. Biochemistry and Biophysical Research Communications 229 (1), 58 – 64.
K. Mackie / Life Sciences 77 (2005) 1667–1673
1673
Sugiura, T., Kodaka, T., Nakane, S., Miyashita, T., Kondo, S., Suhara, Y., Takayama, H., Waku, K., Seki, C., Baba, N., Ishima, Y., 1999. Evidence that the cannabinoid CB1 receptor is a 2-arachidonoylglycerol receptor. Structure-activity relationship of 2-arachidonoylglycerol, ether-linked analogues, and related compounds. Journal of Biological Chemistry 274 (5), 2794 – 2801. Wager-Miller, J., Westenbroek, R., Mackie, K., 2002. Dimerization of G protein-coupled receptors: CB1 cannabinoid receptors as an example. Chemistry and Physics of Lipids 121 (1–2), 83 – 89.