The concept that G protein-coupled receptors (GPCRs) 1 exist as dimers or higher order oligomers has moved rapidly from hypothesis to being widely accepted (1-4). A range of approaches has contributed to this understanding. This includes the ability to co-immunoprecipitate differentially epitopetagged forms of a GPCR from cells in which they are co-expressed, and, in intact cells, the application of a number of resonance energy transfer-based techniques. However, the role of dimerization in function and the mechanisms then responsible for initiation of signal transduction by the dimer have been more recalcitrant to analysis. Significant progress in this area has recently been achieved for the class C GPCRs. These contain both a long extracellular N-terminal domain, to which agonist ligands bind, and the prototypic seven transmembrane (TM) helix bundle architecture that is the common feature of all GPCR families (for review, see Ref. 5). The functional ␥-aminobutyric acid, type b receptor is a hetero-dimer of two distinct gene products in which trafficking to the plasma membrane requires interaction between the partner polypeptides (6 -9). This indicates that a key role of dimerization is achieving appropriate cellular localization. This is also true for the class A rhodopsin-like GPCRs because non-functional, truncated splice variants can restrict plasma membrane delivery of fulllength GPCRs and, thus, limit their function (10 -12). Chimeric class C GPCRs consisting of the extracellular domain of one GPCR and the TM and intracellular elements of a second, closely related GPCR or in which the intracellular loops of dimer partners are exchanged have provided strong evidence that the mechanism of action of the dimer involves transactivation (13)(14); that is, ligand binding to one element of the dimer results in activation of G protein produced by the other GPCR within the dimer. Again, the ␥-aminobutyric acid, type b receptor has been particularly informative in this regard as only one of the two gene products that forms the dimer is able to bind the agonist ␥-aminobutyric acid. Equivalent systems are not available for the majority of class A GPCRs. However, for the luteinizing hormone receptor, which also has a long N-terminal domain that binds the ligand, partial reconstitution of function has been achieved by co-expression of distinct pairs of mutants (15-16). Furthermore, in recent studies co-expression of a mutant receptor defective in hormone binding and another mutant defective in signal generation rescued hormone-activated cAMP production (17).