**5. Ligand binding studies and the ternary complex model**

Ligands that induce G protein activation are termed "agonists", and ligands that do not affect the receptor activity, "antagonists". Even at 100% receptor occupancy, some agonists have a larger effect than others on G protein activation: the more effective agonists are called "full agonists" and the less efficient compounds, "partial agonists". More recently, it has been demonstrated that most GPCRs have the ability to activate (inefficiently) their cognate G proteins in the absence of any ligand: this is called "constitutive activity". Compounds that counteract the receptors' constitutive activity are called "inverse agonists".

Agonist binding to GPCRs in the absence of either GDP or GTP facilitates the formation of the ternary complex involving the receptor, an agonist, and an empty G protein (Figure 3) (Lefkowitz *et al.*, 1976; De Lean *et al.*, 1980; Rasmussen *et al.*, 2011b). This is evident from the formation of a high molecular weight "ternary complex" (ligand-receptor-G protein, LRG) with a much higher affinity for agonists compared to isolated receptors. Guanyl nucleotides (GTP, GTP analogues or GDP) destabilize the G protein interaction with agonist-bound receptors by markedly decreasing the G-protein affinity for the receptor. When recognizing the high affinity ternary complex, guanyl nucleotides dramatically increase the agonists' dissociation rate, and decrease the receptor affinity for agonists while increasing their affinity for inverse agonists (see for instance (Lefkowitz *et al.*, 1976; Berrie *et al.*, 1979)); GDP is typically needed in larger concentrations than GTP or GTP analogues.

The effect of GTP on ligand binding can be used as a measure of the relative stability of the ternary complex compared to the binary ligand-receptor complex (Lefkowitz *et al.*, 1976). Full agonists have a higher affinity in the absence of GTP and inverse agonists have a higher affinity in its presence: the effect of GTP on ligand recognition is correlated with the ligands' ability to induce or inhibit G protein activation by the receptor (Lefkowitz *et al.*, 1976; De Lean *et al.*, 1980).

The original ternary complex model (Figure 7: top left) (Lefkowitz *et al.*, 1976) was designed to describe ligand binding to GPCRs. It describes the allosteric interactions between the ligand (L) and the G protein (G) recognizing different binding sites on the same receptor (R). Guanyl nucleotides were assumed to "prevent" G protein interaction with the receptor. The ternary complex model was later completed to the cubic ternary complex model (Figure 7: bottom right) (Weiss *et al.*, 1996a; Weiss *et al.*, 1996c; Weiss *et al.*, 1996b). Two receptor conformations (R and R\*) without and with the ability to activate G proteins respectively are assumed to coexist at equilibrium (RR\*) in the absence and presence of ligands or G proteins. Agonists and G proteins favor the active (R\*) receptor conformation while inverse agonists stabilize the inactive (R) conformation. As in the ternary complex model, the cubic model describes the binding (as opposed to functional) properties of GPCRs; guanyl nucleotides are assumed to "prevent" the receptor-G protein interaction.

Fig. 7. The ternary complex model. Top left: the ternary complex model assumes that the receptor (R) can interact simultaneously with a ligand (L) and the G protein (G). Agonists facilitate and antagonists inhibit the receptor-G protein interaction; the ternary complex (LRG) is somehow responsible for transduction of the effect. Bottom right: the "cubic ternary complex model" assumes that the receptor can be found in a resting (R) or in an active (R\*) conformation. G proteins (G) and agonist ligands stabilize R\* while inverse agonists stabilize the R conformation. R\*G and LR\*G complexes are responsible for the biological effects of the active receptor.

Both ternary complex models have had a tremendous impact on our vision of GPCR function: the agonist-receptor-G protein complex is more and more often considered as "**the** active receptor". It should be remembered, however, that the ternary complex is certainly not "biologically active": it accumulates only under conditions where the G protein is unable to activate its' effectors (in the absence of GTP), and the G protein conformation in the β-adrenergic receptor-G protein complex (Figure 3) is not compatible with GS-adenylate cyclase interaction (Figure 5)! The ternary complex model was designed to describe ligand binding to the receptors, as opposed to effectors activation.

ligand (L) and the G protein (G) recognizing different binding sites on the same receptor (R). Guanyl nucleotides were assumed to "prevent" G protein interaction with the receptor. The ternary complex model was later completed to the cubic ternary complex model (Figure 7: bottom right) (Weiss *et al.*, 1996a; Weiss *et al.*, 1996c; Weiss *et al.*, 1996b). Two receptor conformations (R and R\*) without and with the ability to activate G proteins respectively are assumed to coexist at equilibrium (RR\*) in the absence and presence of ligands or G proteins. Agonists and G proteins favor the active (R\*) receptor conformation while inverse agonists stabilize the inactive (R) conformation. As in the ternary complex model, the cubic model describes the binding (as opposed to functional) properties of GPCRs; guanyl

Fig. 7. The ternary complex model. Top left: the ternary complex model assumes that the receptor (R) can interact simultaneously with a ligand (L) and the G protein (G). Agonists facilitate and antagonists inhibit the receptor-G protein interaction; the ternary complex (LRG) is somehow responsible for transduction of the effect. Bottom right: the "cubic ternary complex model" assumes that the receptor can be found in a resting (R) or in an active (R\*) conformation. G proteins (G) and agonist ligands stabilize R\* while inverse agonists stabilize the R conformation. R\*G and LR\*G complexes are responsible for the

Both ternary complex models have had a tremendous impact on our vision of GPCR function: the agonist-receptor-G protein complex is more and more often considered as "**the** active receptor". It should be remembered, however, that the ternary complex is certainly not "biologically active": it accumulates only under conditions where the G protein is unable to activate its' effectors (in the absence of GTP), and the G protein conformation in the β-adrenergic receptor-G protein complex (Figure 3) is not compatible with GS-adenylate cyclase interaction (Figure 5)! The ternary complex model was designed to describe ligand

biological effects of the active receptor.

binding to the receptors, as opposed to effectors activation.

nucleotides are assumed to "prevent" the receptor-G protein interaction.
