Figure 4: Quantifying the Gi bias of dualsteric probes.
From: The allosteric vestibule of a seven transmembrane helical receptor controls G-protein coupling
(a) M2 receptor-mediated Gi protein activation is reflected by [35S]GTPγS binding to membranes of CHO-hM2 cells. Maximum G-protein activation induced by ACh was set 100%. Data are means±s.e.m. from at least four independent experiments conducted in quadruplicates. (b) M2 receptor-mediated Gs activation is reflected by an increase of intracellular cAMP in CHO-hM2 cells pretreated with PTX (50 ng ml−1, 16–24 h). Maximum cAMP formation induced by ACh was set 100%. Data are means±s.e.m. from at least three independent experiments conducted in quadruplicates. Concentration-effect curves were obtained by global fits applying the operational model of agonism37 to the data points in a and b. (c) Visualization of ligand bias: fractional [35S]GTPγS binding and cAMP accumulation (a,b) are plotted for equal concentrations of the respective agonists on the ordinate and abcissa, respectively. The hyperbolic shape of the ACh graph indicates system bias. The Gi protein is preferred by the M2 receptor. Ligands whose bias exceeds system bias display graphs that differ substantially in their curvature radii and the height of their plateaus. (d) Quantification of ligand bias: ΔΔlog(τ/KA) values compensate for system bias and display true ligand bias relative to Ach, which is taken as balanced agonist. Ligands with positive values are Gi biased when the limits of their 95% confidence intervals (CIs) do not contain zero. Data are means with 95% CIs. (e) Efficacies of the agonist-occupied receptor to activate the Gi pathway (log τ(Gi)) and the Gs pathway (log τ(Gs)) are plotted on the ordinate and abcissa, respectively. Log τ values are derived from global fits applying the operational model to the data points in a and b. Data are means±s.e.m.
