a, 24-hour pre-treatment with 100 ng/mL PTX abolishes high-affinity binding (≤ 1 %) due to Gαi inactivation. Wt- data from Supplementary Fig. 1 (n = 3 independent experiments). The PTX curve is a representative one of 3 independent experiments with similar results, datapoints of the curve are mean ± SD. b, Fraction of high- affinity binding is dramatically increased for the constitutively active T279K mutant (64.4 ± 1.6 %). High- affinity binding is abolished in the T279D mutant (≤ 1 %), comparable to PTX treated wild-type receptor. Data are mean ± SEM from n = 3 independent experiments of each condition. P values were determined by one-way ANOVA and Tukey’s multiple comparison test. c, For the T279K mutant G protein activation occurs already without DAMGO stimulation, leading to a full response of the Gi- FRET sensor in the basal state. The response of the T279D mutant is right-shifted to the wild-type (pEC50-wt= 9.2 ± 0.1 vs. pEC50-T279D = 6.6 ± 0.1). Data are mean ± SEM from n = 4 independent experiments of each variant. d, Measurement of basal receptor activity and its mutants, based on Gi-mediated cAMP decrease, after receptor activation with DAMGO. Cells were pre-stimulated with 1 µM forskolin to slightly elevate cAMP levels. The pEC50 of the constitutively active mutant (10.1 ± 0.1) does not significantly differ from µOR-wild-type (9.9 ± 0.4). Due to the high basal activity of the T279K mutant the amplitude of additional Gi activation via DAMGO is dramatically reduced (20.8 ± 1.7 %), compared to the wild-type receptor (100 %). The inactive T279D mutant shows a pEC50 right-shifted by 2–3 log-units (7.4 ± 0.1) while the amplitude is slightly but not significantly higher than the one of the wild-type receptor (103 ± 2 %). Data from n = 4 independent experiments. Numbers are means and errors are given as SEM.