Effector membrane translocation biosensors reveal G protein and βarrestin coupling profiles of 100 therapeutically relevant GPCRs

  1. Charlotte Avet
  2. Arturo Mancini
  3. Billy Breton
  4. Christian Le Gouill
  5. Alexander S Hauser
  6. Claire Normand
  7. Hiroyuki Kobayashi
  8. Florence Gross
  9. Mireille Hogue
  10. Viktoriya Lukasheva
  11. Stéphane St-Onge
  12. Marilyn Carrier
  13. Madeleine Héroux
  14. Sandra Morissette
  15. Eric B Fauman
  16. Jean-Philippe Fortin
  17. Stephan Schann
  18. Xavier Leroy  Is a corresponding author
  19. David E Gloriam  Is a corresponding author
  20. Michel Bouvier  Is a corresponding author
  1. Institute for Research in Immunology and Cancer (IRIC), and Department of Biochemistry and Molecular Medicine, Université de Montréal, Canada
  2. Domain Therapeutics North America, Canada
  3. Department of Drug Design and Pharmacology, University of Copenhagen, Denmark
  4. Internal Medicine Research Unit, Pfizer Worldwide Research, Development and Medical, United States
  5. Pfizer Global R&D, United States
  6. Domain Therapeutics, France
9 figures, 3 videos, 1 table and 6 additional files

Figures

EMTA ebBRET platform to monitor G protein activation and βarrestin recruitment.

(A) Schematic of the G protein Effector Membrane Translocation Assay (GEMTA) to monitor Gα protein activation. Upon receptor activation, RlucII-tagged effector proteins (Effector-RlucII) translocate towards and interact with active Gα subunits from each G protein family, leading to increased ebBRET. (B) Principle of the Effector Membrane Translocation Assay (EMTA) monitoring βarrestin recruitment to the plasma membrane (top) and Gαs activation (bottom). Top; upon receptor activation, RlucII-tagged βarrestins (βarrestin-RlucII) translocate to the plasma membrane, thus increasing ebBRET with rGFP-CAAX. Bottom; Internalization of activated RlucII-tagged Gαs (Gαs-RlucII) following receptor stimulation decreases ebBRET with the membrane-anchored rGFP-CAAX.

Figure 2 with 6 supplements
Validation of EMTA ebBRET-based platform to monitor Gα protein activation.

(A) Pharmacological validation of the Gαi/o activation sensor. HEK293 cells were transfected with the D2 receptor and the Gαi/o family-specific sensor, along with each Gαi/o subunit. Concentration-response curve using the Gαi/o activation sensor, in the presence or absence of UBO-QIC (left) or PTX (right) inhibitors. Insets; Emax values determined from concentration-response curves of inhibitor-pretreated cells. (B) Pharmacological validation of the Gαq/11 activation sensor. HEK293 cells were transfected with the GnRH receptor and the Gαq/11 family-specific sensor, along with each Gαq/11 subunit. Concentration-response curve using Gαq/11 activation sensor, in the presence or absence of UBO-QIC (left) or PTX (right) inhibitors. Insets; Emax values determined from dose-response curves of inhibitor-pretreated cells. (C) Validation of the Gα12/13 activation sensor. Cells were transfected with the CB1 receptor and one of the Gα12/13 activation sensors, along with the Gα12 or Gα13 subunits. Concentration-response curves of HEK293 cells (top) or the parental and devoid of G12/13 (ΔG12/13) HEK293 cells (bottom) using the PDZ-RhoGEF-RlucII/rGFP-CAAX (top and bottom left) or PKN-RBD-RlucII/rGFP-CAAX (bottom right) sensors, pretreated or not with UBO-QIC or PTX (top). (D) Pharmacological validation of the Gαs activation sensor. HEK293 cells were transfected with the GPBA receptor and the Gαs activation (left and central) or the EPAC (right) sensors. Left: Concentration-response curves using the Gαs activation sensor in the presence or absence of UBO-QIC or PTX, inhibitors of Gαq or Gαi/o, respectively. Central: Concentration-response activation of the Gαs sensor using CTX, a Gαs activator. Right: Concentration-response curve using the EPAC sensor. Inset; Emax values determined from dose-response curves of inhibitors-pretreated cells. Data are expressed as BRET ratio for the concentration-response curves or expressed in % of respective control cells (Emax graphs) and are the mean ± SEM of 3 (A–C) or 4 (D) independent experiments performed in one replicate. Unpaired t-test (A–D): *p < 0.05 and ***p < 0.001 compared to control cells.

Figure 2—figure supplement 1
Influence of endogenous G proteins.

Concentration-response curves elicited in parental (WT) HEK293 cells or devoid of Gs (ΔGs), G12/13 (ΔG12/13), Gq/11 (ΔGq/11), or Gi/o (ΔGi/o) proteins, transfected with the indicated receptor (D2, GnRHR, GIP, V2, EP3, or M3) and one of the Gαi/o, Gαq/11, or Gα12/13 activation sensors, along with the indicated Gα subunits. Mock condition corresponded to the response elicited in absence of heterologously expressed Gα subunits (i.e. endogenous G proteins effect). Data are the mean ± SEM of 3 -5 independent experiments performed in one replicate and are expressed as BRET2 ratio. Data presented in (I) are the same as in (A–B), but with results expressed as % of maximal response elicited by endogenous G proteins (mock) in WT cells.

Figure 2—figure supplement 2
Validation of EMTA ebBRET-based sensors selectivity for each Gα subunit families.

HEK293 cells were transfected with the ETA receptor and Gαi/o (A), Gαq/11 (B), or Gα12/13 (C) activation sensors along with each Gα subunit or control DNA (Mock) as control for response obtained with endogenous Gα proteins. Concentration-response curves in response to endothelin-1 are shown (left and central), as well as maximal responses obtained with each Gα subunit. Data are the mean ± SEM of 3 independent experiments performed in one replicate and are expressed as BRET2 ratio. Unpaired t-test: #p < 0.05, ##p < 0.01 and ###p < 0.001 compared to Mock (without receptor) and one-way ANOVA test: *p < 0.05, **p < 0.01 and ***p < 0.001 compared to Mock + ETA.

Figure 2—figure supplement 3
Influence of G protein, GPCR or effector-RlucII level expression.

(A) Concentration-response curves elicited in HEK293 cells transfected with the B2 receptor and one of the Gαq/11, Gαi/o, or Gα12/13 activation sensors, along with increasing quantity of the indicated Gα subunits. Data represent a representative experiment (B) Concentration-response curves elicited in HEK293 cells transfected with increasing quantity of the M3, D2, or AT1 receptors and the Gαq/11, Gαi/o, or Gα12/13 activation sensors, along with the indicated Gα subunits. (C) Concentration-response curves elicited in HEK293 cells transfected with the ETA receptor and increasing quantity of effector-RlucII (p63-RhoGEF for Gαq/11, Rap1GAP for Gαi/o or PDZ-RhoGEF for Gα12/13), along with rGFP-CAAX and the indicated Gα subunits. Data are the mean ± SEM of 3 independent experiments performed in one replicate and are expressed in BRET2 ratio.

Figure 2—figure supplement 4
Kinetics of Gα proteins and βarrestins recruitment promoted by the ETA receptor.

Kinetics of activation of the indicated pathways following stimulation with vehicle or Endothelin-1 in HEK293 cells expressing the ETA receptor. Data are the mean ± SD of two replicates of a representative experiment from 3 independent experiments and are expressed in % of the respective basal response (determined before ligand addition at t = 0 sec).

Figure 2—figure supplement 5
Comparison of EMTA platform and G protein activation BRET assay based on Gαβγ dissociation.

Concentration-response curves elicited in HEK293 cells transfected with the indicated receptor (D2, GIP, PTH1, M3, ETA, B1, FP, or Cys-LT2) and one of the Gαi/o, Gαq/11, or Gα12/13 EMTA activation sensors, along with the indicated Gα subunits, or the BRET-based Gαβγ dissociation sensors (Gα-RlucII and GFP10-Gγ1 for Gαq, Gα12, and Gα13 or GFP10-Gγ2 for Gαi1, Gαi2, and GαoB, with untagged Gβ1). Data are the mean ± SEM from 3-7 independent experiments performed in one replicate and results are expressed in % of the response obtained for cells treated with vehicle.

Figure 2—figure supplement 6
Western blots of G protein level expression in cells transfected with the EMTA ebBRET platform.

G protein expression level detection in HEK293 cells transfected with the Gαi/o, Gα12/13, Gαq/11, or Gαs activation sensors along with the indicated Gα protein or control DNA (Mock). Representative immunoblots of 3 independent experiments are shown.

Figure 3 with 3 supplements
Heatmaps illustrating the diversity of receptor-specific signaling signatures detected with the EMTA ebBRET platform.

(A) First, values within each pathway were normalized relative to the maximal response observed across all receptors (max = 1; left). These values were then normalized across pathways for the same receptor, with the highest-ranking pathway serving as the reference (max = 1; right). (B) Heatmap representation of double normalized Emax (left) and pEC50 (right) data. Empty cells (grey) indicate no detected coupling. IUPHAR receptor names are displayed.

Figure 3—figure supplement 1
Receptor-specific signaling signatures.

Emax values derived from concentration-response curves generated on 100 receptors using the 15 ebBRET-based assay are represented as radial graphs. A score of 0 indicates no coupling to a given pathway, whereas a score of 1 indicates a coupling. Receptors are rearranged according to the number of G protein families activated. Gα15 has been considered apart from the Gq/11 family due to its promiscuous nature. See Supplementary file 3 that shows the concentration-response curves of the 100 receptors for the 15 different pathways.

Figure 3—figure supplement 2
Detection of endogenous receptor-mediated responses with the EMTA ebBRET platform in HEK293 cells.

Comparison of concentration-response curves elicited by the indicated ligand for a specific pathway, following the stimulation of HEK293 cells expressing endogenous or heterologously expressed receptors. The data presented refer to the ligands where a signal was detected on non-transfected cells (endogenous expression) (See Supplementary file 3 for the curves on light gray and yellow background). Data are the mean ± SEM of at least 3 independent experiments performed in one replicate and expressed in % of the response obtained for cells treated with vehicle.

Figure 3—figure supplement 3
Validation of G12/13 and G15 signaling for the newly characterized GPCRs.

(A) Validation of G12/13-mediated signal using Rho and Ezrin activation sensors. HEK293 cells expressing FP or CysLT2 receptors and the PKN-RBD-RlucII or MyrPB-Ezrin-RlucII/rGFP-CAAX sensors were pretreated or not with the Gαq inhibitor YM-254890 and then stimulated with increasing concentrations of respective ligand. Data are the mean ± SEM from 3-5 independent experiments performed in one replicate and expressed in % of vehicle-treated cells. (B) Validation of Gα15-mediated signal by measuring calcium production. Top: Kinetics of calcium release induced by the indicated ligand in HEK293 cells expressing the indicated receptor, alone or with Gα15 subunit. For receptors that also couple to other Gq/11 family members, cells were pretreated with DMSO or the Gαq inhibitor YM-254890. Bottom: The peak of calcium production obtained from kinetics were compared to the basal level of calcium (determined between 0 and 17 s). Data are the mean ± SEM from 5-7 independent experiments performed in one replicate and expressed in relative fluorescence unit (RFU). Two Way ANOVA test: *p < 0.05, **p < 0.01 and ***p < 0.001 compared to respective basal calcium level. ns: not significant.

Figure 4 with 1 supplement
The EMTA ebBRET platform has a unique ability to uncover coupling selectivity between G protein families.

(A) Venn diagram showing the numbers of receptors coupled to each G protein family in the EMTA ebBRET biosensor assay. (B) Evaluation of receptors coupling promiscuity: number of receptors that couple to members of 1, 2, 3, or 4 G protein families. (C) Determination of G protein subunit coupling frequency: number of receptors that activate each Gα subunit. (D) Proportion of receptors recruiting βarrestins: number of receptors that do not recruit (-/-) or that recruit either (+/- or -/+) or both (+/+) βarrestin isotypes. All data are based on double normalized Emax values from Figure 3.

Figure 4—figure supplement 1
G protein subtypes distribution across the 100 GPCRs profiled with the EMTA ebBRET-based platform.

(A) Number of receptors that can couple to 1–5 of the different subtypes from each G protein family. (B) % of receptors activating a specific G protein subtype (Y axis) that also activate another G protein subtype (X axis).

Figure 5 with 1 supplement
Multiple applications using the EMTA ebBRET platform.

(A) Inverse agonist activity detection. Left: Gαi2 activation in HEK293 cells transfected with the Rap1GAP-RlucII/rGFP-CAAX sensors with untagged Gαi2 and increasing amount of A1 receptor plasmid. Data are expressed in % of response obtained in control cells (0 ng of A1) and are the mean ± SEM of 4–6 independent experiments performed in two replicates. One Way ANOVA test: ***p < 0.001 compared to control cells. HEK293 cells expressing the Gαi2 activation sensor and control (Mock) or A1 receptor plasmid were stimulated (10 min) with increasing concentrations of the indicated compound. Data are expressed in % of constitutive response obtained in vehicle-treated A1 transfected cells and are the mean ± SEM of 4-6 independent experiments performed in one replicate. Right:z activation in HEK293 cells transfected with the Rap1GAP-RlucII/rGFP-CAAX sensors with untagged Gαz and increasing amount of CB1 receptor plasmid. Data are expressed in % of response obtained in control cells (0 ng of CB1) and are the mean ± SEM of 4 independent experiments performed in one replicate. One Way ANOVA test: ***p < 0.001 compared to control cells. HEK293 cells expressing the Gαz activation sensor and increasing amount of CB1 receptor plasmid were directly stimulated (10 min) with increasing concentrations of the CB1 inverse agonist rimonabant. Data are expressed as % of the response obtained in control cells (0 ng of CB1) treated with vehicle and are the mean ± SEM of 4 independent experiments performed in one replicate. (B) Ligand-biased detection. Concentration-response curves of AT1 for the endogenous ligand (Angiotensin II, AngII) and biased agonists [Sar1-Ile4-Ile8] AngII (SII), saralasin or TRV027. G protein and βarrestin2 signaling activity were assessed by EMTA platform. Data are expressed in % of maximal response elicited by AngII and are the mean ± SEM of 3–6 independent experiments performed in one replicate. (C) Functional selectivity of naturally occurring receptor variants. Concentration-response curves for WT or E/DRY motif Asp128Asn and Arg129His variants of GPR17 upon agonist stimulation in HEK293 cells co-expressing the indicated EMTA biosensor. Data are expressed in % of maximal response elicited by WT receptor and are the mean ± SEM of 3 independent experiments performed in one replicate.

Figure 5—figure supplement 1
Modulation of ligand-promoted response detected by EMTA ebBRET platform by receptor constitutive activity.

(A) Concentration-response curves of Gαi2 activation elicited by adenosine in HEK293 cells transfected with the Rap1GAP-RlucII/rGFP-CAAX sensors with untagged Gαi2 and A1 or A3 receptors. Basal level of Gi2 activation detected by the GEMTA sensor in absence of heterologous receptor expression is represented by the interrupted line. Data are expressed as uBRET ratio and are the mean ± SEM of 4 independent experiments performed in one replicate. (B) Concentration-response curves of Gαq activation elicited by serotonin in HEK293 cells transfected with the p63-RlucII/rGFP-CAAX sensors with untagged Gαq and increasing amount of 5-HT2C receptor plasmid. Data are expressed as BRET ratio and are the mean ± SEM of 4 independent experiments performed in one replicate.

Figure 6 with 2 supplements
Detection of direct and indirect (trans) mechanisms of ligand polypharmacology using the Gz/G15 biosensor.

(A) Test of the Gz/G15 biosensor on a safety target panel. ebBRET signal was measured before and after stimulation with the indicated ligand in HEK293 cells transfected with the combined Gz/G15 biosensor and one of the 24 receptors listed. (B) Cross-activation of D2 and α2AAR by other natural ligands. For the agonist mode read, HEK293 cells expressing D2 or α2AAR and either the Gαi2, GαoB, or the βarrestin2 + GRK2 sensors were stimulated with increasing concentrations of the indicated ligand. For the antagonist mode read, cells were pretreated with increasing concentrations of the selective D2 antagonist eticlopride or the selective α2AAR antagonist WB4101 before stimulation with an EC80 of the indicated ligand. Data are the mean ± SEM from 3-4 independent experiments performed in one replicate and expressed in % of the response elicited by dopamine or noradrenaline for D2 and α2AAR expressing cells, respectively. (C) Indirect (trans) activation of CB1 by acetylcholine. For the agonist mode read, HEK293 cells expressing CB1 and the Rap1GAP-RlucII/rGFP-CAAX sensors with untagged GαoB were stimulated with increasing concentrations of the indicated ligand. For the antagonist mode read, same cells were pretreated or not with increasing concentrations of the CB inverse agonist AM-630 (left) or the cholinergic antagonist atropine (central) before stimulation with an EC80 of the indicated ligand. To evaluate the contribution of Gq/11-coupled receptor, cells were pretreated with the Gαq inhibitor UBO-QIC and then stimulated with increasing concentrations of the indicated ligand (right). Data are the mean ± SEM from 3-5 independent experiments performed in one replicate and expressed in % of the response elicited by WIN55,212–2.

Figure 6—figure supplement 1
Combined Gz/G15 biosensor.

HEK293 cells transfected with the Rap1GAP-RlucII/p63-RhoGEF-RlucII/rGFP-CAAX sensors along with Gαz and Gα15 subunits and the indicated untagged receptor were stimulated with increasing concentrations of the indicated ligand. Data are the mean ± SEM from 3-5 independent experiments performed in one replicate and results are expressed in % of vehicle-treated cells.

Figure 6—figure supplement 2
Validation of direct activation of α2AAR by dopamine.

HEK293 cells expressing D2 or α2AAR and the Gαi2 (A) or the GαoB (B) sensors were pretreated or not with the selective D2-family antagonist eticlopride, before stimulation with increasing concentrations of dopamine. Data are the mean ± SEM from 2-4 independent experiments performed in one replicate and expressed in % of the response elicited by dopamine.

Detection of endogenous receptor- and/or G protein-mediated responses in cells with the EMTA ebBRET platform.

Concentration-dependent activation of Gαi2 protein by (A) endogenous S1P1 receptor in iPSC-derived cardiomyocytes transfected with heterologous Gαi2, (B) endogenous FPR2 in promyelocytic HL-60 cells transfected with heterologous Gαi2, (C) endogenous FPR2 in promyelocytic HL-60 cells with endogenous Gi/o proteins and (D) endogenous PAR2 receptor in HEK293 cells with endogenous Gi/o proteins. In all cases, cells were co-transfected with the Rap1GAP-RlucII/rGFP-CAAX biosensor. Data are the mean ± SEM of 3-4 independent experiments performed in one replicate and are expressed as BRET2 ratio in percentage of response induced by vehicle.

Author response image 1
Author response image 2

Videos

Video 1
BRET-based imagery of p63-RhoGEF-RlucII recruitment to the plasma membrane upon AT1 activation.

HEK293 cells expressing the p63-RhoGEF-RlucII/rGFP-CAAX sensors with Gαq and AT1 were stimulated with Angiotensin II. BRET levels (the ratio of the acceptor photon count to the total photon count) are expressed as a color code (lowest being black and purple, and highest being red and white).

Video 2
BRET-based imagery of Rap1GAP-RlucII recruitment to the plasma membrane upon D2 activation.

HEK293 cells expressing the Rap1GAP-RlucII/rGFP-CAAX sensors with Gαi2 and D2 were stimulated with dopamine. BRET levels (the ratio of the acceptor photon count to the total photon count) are expressed as a color code (lowest being black and purple, and highest being red and white).

Video 3
BRET-based imagery of PDZ-RhoGEF-RlucII recruitment to the plasma membrane upon TPαR activation.

HEK293 cells expressing the PDZ-RhoGEF-RlucII/rGFP-CAAX + Gα13 and TPαR were stimulated with U46619. BRET levels (the ratio of the acceptor photon count to the total photon count) are expressed as a color code (lowest being black and purple, and highest being red and white).

Tables

Appendix 1—key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Cell line (Homo-sapiens)HEK29310.1038/ncomms12178 (Namkung et al., 2016)HEK293 clonal cell line (HEK293SL cells)
Cell line (Homo-sapiens)ΔGs HEK293 cellsDr. A. Inoue (Tohoku University, Sendai, Miyagi, Japan) 10.1124 /mol.116.106419 (Stallaert et al., 2017)HEK293 cells devoid of functional Gαs
Cell line (Homo-sapiens)ΔG12/13 HEK293 cellsDr. A. Inoue (Tohoku University, Sendai, Miyagi, Japan) 10.1074/jbc.M116.763854 (Devost et al., 2017)HEK293 cells devoid of functional Gα12 and Gα13
Cell line (Homo-sapiens)ΔGq/11 HEK293 cellsDr. A. Inoue (Tohoku University, Sendai, Miyagi, Japan) 10.1038/ncomms10156 (Schrage et al., 2015)HEK293 cells devoid of functional Gαq, Gα11, Gα14 and Gα15
Cell line (Homo-sapiens)ΔGi/o HEK293 cellsDr. A. Inoue (Tohoku University, Sendai, Miyagi, Japan)HEK293 cells devoid of functional Gαi and Gαo
Cell line (Homo-sapiens)HL-60ATCCCat. #: CCL-240
Cell line (Homo-sapiens)iCell Cardiomyocytes, 01434FUJIFILM Cellular DynamicsCat. #: R1057
Transfected construct (Homo sapiens)Human Gα subunits-encoding plasmid libraryMissouri S&T cDNA Resource Center (https://www.cdna.org/)Cat. #: GNAI100000; GNAI200000; GNA0OA0000; GNA0OB0000; GNA0Z00000; GNA1200000; GNA1300001; GNA0Q00000; GNA1100000; GNA1400000; GNA1500000; GNA0SL0000
Transfected construct (Homo sapiens)Gβ1Missouri S&T cDNA Resource Center (https://www.cdna.org/)Cat. #: GNB0100000
Transfected construct (Homo sapiens)Gγ9Missouri S&T cDNA Resource Center (https://www.cdna.org/)Cat. #: GNG0900000
Transfected construct (Homo sapiens)Gαs-67-RlucII10.1074/jbc.M114.618819(Carr et al., 2014)
Transfected construct (Homo sapiens)Gαi1-loop-RlucII10.1096/fj.13242446 (Armando et al., 2014)
Transfected construct (Homo sapiens)Gαi2-loop-RlucII10.1073/pnas.1312515110 (Quoyer et al., 2013)
Transfected construct (Homo sapiens)GαoB-99-RlucII10.1073/pnas.1804003115(Mende et al., 2018)
Transfected construct (Homo sapiens)Gαq-118-RlucII10.1016 /j.bpj.2010.10.025 (Breton et al., 2010)
Transfected construct (Homo sapiens)Gα12–136-RlucII10.1126/scisignal.aat1631(Namkung et al., 2018)
Transfected construct (Homo sapiens)Gα13–130-RlucII10.1038 /s42003-020-01453-8 (Avet et al., 2020)
Transfected construct (Homo sapiens)GFP10-Gγ110.1096/fj.13242446 (Armando et al., 2014)
Transfected construct (Homo sapiens)GFP10-Gγ210.1038/nsmb1134(Galés et al., 2006)
Transfected construct (Homo sapiens)EPAC10.1124/jpet.109.156398(Leduc et al., 2009)
Transfected construct (Homo sapiens)rGFP-CAAX10.1038/ncomms12178(Namkung et al., 2016)
Transfected construct (Homo sapiens)Rap1GAP-RlucIIThis paperSee Materials and Methods
Transfected construct (Homo sapiens)p63-RhoGEF-RlucIIThis paperSee Materials and Methods
Transfected construct (Homo sapiens)PDZ-RhoGEF-RlucIIThis paperSee Materials and Methods
Transfected construct (Homo sapiens)PKN-RBD-RlucII10.1126/scisignal.aat1631(Namkung et al., 2018)
Transfected construct (Homo sapiens)MyrPB-Ezrin-RlucII10.1242/jcs.255307(Leguay et al., 2021)
Transfected construct (Homo sapiens)βarrestin1-RlucII10.1126/scisignal.2002522(Zimmerman et al., 2012)
Transfected construct (Homo sapiens)βarrestin2-RlucII10.1073/pnas.1312515110 (Quoyer et al., 2013)
Transfected construct (Homo sapiens)GRK2This paperSee Materials and Methods
Transfected construct (Homo sapiens)RAMP3Domain Therapeutics North AmericaN/A
Transfected construct (Homo sapiens)EAAC-110.1016/s0028-3908(98)00091-4 (Brabet et al., 1998)
Transfected construct (Homo sapiens)EAAT-1Domain Therapeutics North AmericaN/A
Transfected construct (Homo sapiens)signal peptide-Flag-AT110.1074/jbc.M114.631119(Goupil et al., 2015)
Transfected construct (Homo sapiens)FLAG-α2BARDomain Therapeutics North AmericaN/A
Transfected construct (Homo sapiens)HA-β2AR10.1074/jbc.M204163200(Lavoie et al., 2002)
AntibodyGαi1 (I-20) (Rabbit polyclonal)Santa CruzCat. #: sc-391 RRID: AB_2247692WB (1:500)
AntibodyGαi2 (T-19) (Rabbit polyclonal)Santa CruzCat. #: sc-7276 RRID:AB_2111472WB (1:500)
AntibodyGαo (K-20) (Rabbit polyclonal)Santa CruzCat. #: sc-387 RRID:AB_2111641WB (1:500)
AntibodyGαz (Rabbit monoclonal)AbcamCat. #: ab154846WB (1:1000)
AntibodyGαs (K-20) (Rabbit polyclonal)Santa CruzCat. #: sc-823 RRID:AB_631538WB (1:500)
AntibodyGα12 (S-20) (Rabbit polyclonal)Santa CruzCat. #: sc-409 RRID:AB_2263416WB (1:500)
AntibodyGα13 (A-20) (Rabbit polyclonal)Santa CruzCat. #: sc-410 RRID:AB_2279044WB (1:500)
AntibodyGαq (E-17) (Rabbit polyclonal)Santa CruzCat. #: sc-393 RRID:AB_631536WB (1:500)
AntibodyGα11 (C-terminal) (Rabbit polyclonal)Sigma-AldrichCat. #: SAB2109181WB (1:500)
AntibodyGα14 (Rabbit polyclonal)Sigma-AldrichCat. #: SAB4300771WB (1:500)
AntibodyGα15 (Rabbit polyclonal)ThermoFisher scientific (Pierce)Cat. #: PA1-29022 RRID:AB_1958024WB (1:5,000)
Antibodyβactin (Mouse monoclonal)Sigma-AldrichCat. #: A5441 RRID:AB_476744WB (1:5,000)
AntibodyAnti-rabbit HRP-coupled (Donkey polyclonal)GE HealthcareCat. #: NA934 RRID:AB_772206WB (1:5,000)
AntibodyAnti-mouse HRP-coupled (Sheep polyclonal)GE HealthcareCat. #: NA931 RRID:AB_772210WB (1:10,000)
Commercial assay or kitFLIPR Calcium 5 Assay KitMolecular DevicesCat. #: R8185
Chemical compound, drugα-linolenic acidCayman ChemicalCat. #: 21,910
Chemical compound, drugα-MSHSigma-AldrichCat. #: M4135
Chemical compound, drugγ-MSHTocrisCat. #: 4,272
Chemical compound, drug[Pyr1]-Apelin 13TocrisCat. #: 2,420
Chemical compound, drug[Sar1, Ile4,8]-Angiotensin IIPeptides InternationalCat. #: PAN-4476-V-1EA
Chemical compound, drug3-hydroxyoctanoic acid (3-HOA)Sigma-AldrichCat. #: H3898
Chemical compound, drug7α–25 dihydroxycholesterolSigma-AldrichCat. #: SML0541
Chemical compound, drugAcetylcholine chlorideTocrisCat. #: 2,809
Chemical compound, drugACT-389949Provided by Bristol-Myers SquibbN/A
Chemical compound, drugAdenosineSigma-AldrichCat. #: A9251
Chemical compound, drugAM-630TocrisCat. #: 1,120
Chemical compound, drugAmylinTocrisCat. #: 3,418
Chemical compound, drugAngiotensin II (Ang II)Sigma-AldrichCat. #: A9525
Chemical compound, drugArginine vasopressin (AVP)Sigma-AldrichCat. #: V9879
Chemical compound, drugAtropineSigma-AldrichCat. #: A0132
Chemical compound, drugBovine serum albuminSigma-AldrichCat. #: A7030
Chemical compound, drugC5aComplement TechnologyCat. #: A144(300)
Chemical compound, drugCalcitoninBachemCat. #: H-2250
Chemical compound, drugCCK Octapeptide, sulfated (CCK8)TocrisCat. #: 1,166
Chemical compound, drugCCL20R&D SystemsCat. #: 360-MP/CF
Chemical compound, drugCCL3 (MIP-1a)R&D SystemsCat. #: 270-LD/CF
Chemical compound, drugCholera Toxin (CTX) from Vibrio choleraeSigma-AldrichCat. #: C8052
Chemical compound, drugCmpd43Provided by Bristol-Myers SquibbN/A
Chemical compound, drugCorticotropin-Releasing Factor (CRF)BachemCat. #: H-2435
Chemical compound, drugCXCL12R&D SystemsCat. #: 350-NS
Chemical compound, drugCXCL13R&D SystemsCat. #: 801 CX/CF
Chemical compound, drugCXCL8R&D SystemsCat. #: 208-IL/CF
Chemical compound, drugDAMGOTocrisCat. #: 1,171
Chemical compound, drugDopamineSigma-AldrichCat. #: H8502
Chemical compound, drugDPCPXTocrisCat. #: 0439
Chemical compound, drugDynorphin ATocrisCat. #: 3,195
Chemical compound, drugEndothelin-1TocrisCat. #: 1,160
Chemical compound, drugEticloprideTocrisCat. #: 1,847
Chemical compound, drugFingolimodProvided by Bristol-Myers SquibbN/A
Chemical compound, drugGastric Inhibitory Peptide (GIP)BachemCat. #: H-5645
Chemical compound, drugGhrelinTocrisCat. #: 1,463
Chemical compound, drugGlucagon (Aittaleb et al., 2010; Aittaleb et al., 2011; Armando et al., 2014; Atwood et al., 2011; Avet et al., 2020; Azzi et al., 2003; Bowes et al., 2012; Brabet et al., 1998; Breton et al., 2010; Bünemann et al., 2003; Carr et al., 2014; Casey et al., 1990; Chandan et al., 2021; De Haan and Hirst, 2004; Devost et al., 2017; Fukuhara et al., 2001; Galandrin et al., 2007; Galés et al., 2005; Galés et al., 2006; Goupil et al., 2015; Hauser et al., 2017; Hauser et al., 2022; Hoffmann et al., 2005; Inoue et al., 2019; Jordan et al., 1999; Kawamata et al., 2003; Kenakin, 2019; Kim et al., 2002)BachemCat. #: H-6790
Chemical compound, drugGlucagon-like peptide-1 GLP-1 (Bowes et al., 2012; Brabet et al., 1998; Breton et al., 2010; Bünemann et al., 2003; Carr et al., 2014; Casey et al., 1990; Chandan et al., 2021; De Haan and Hirst, 2004; Devost et al., 2017; Fukuhara et al., 2001; Galandrin et al., 2007; Galés et al., 2005; Galés et al., 2006; Goupil et al., 2015; Hauser et al., 2017; Hauser et al., 2022; Hoffmann et al., 2005; Inoue et al., 2019; Jordan et al., 1999; Kawamata et al., 2003; Kenakin, 2019; Kim et al., 2002; Kobayashi et al., 2019; Laschet et al., 2019; Lavoie et al., 2002; Leduc et al., 2009; Leguay et al., 2021; Lu et al., 2014; Lutz et al., 2007)BachemCat. #: H-6795
Chemical compound, drugGlucagon-like peptide-2 GLP-2 (Aittaleb et al., 2010; Aittaleb et al., 2011; Armando et al., 2014; Atwood et al., 2011; Avet et al., 2020; Azzi et al., 2003; Bowes et al., 2012; Brabet et al., 1998; Breton et al., 2010; Bünemann et al., 2003; Carr et al., 2014; Casey et al., 1990; Chandan et al., 2021; De Haan and Hirst, 2004; Devost et al., 2017; Fukuhara et al., 2001; Galandrin et al., 2007; Galés et al., 2005; Galés et al., 2006; Goupil et al., 2015; Hauser et al., 2017; Hauser et al., 2022; Hoffmann et al., 2005; Inoue et al., 2019; Jordan et al., 1999; Kawamata et al., 2003; Kenakin, 2019; Kim et al., 2002; Kobayashi et al., 2019; Laschet et al., 2019; Lavoie et al., 2002; Leduc et al., 2009)BachemCat. #: H-7742
Chemical compound, drugGlutamateSigma-AldrichCat. #: 49,621
Chemical compound, drugGnRH (LH-RH)Peptides InternationalCat. #: PLR-4013
Chemical compound, drugHistamineTocrisCat. #: 3,545
Chemical compound, drugKallidinAnaspecCat. #: 22,853(AN)
Chemical compound, drugLeukotriene B4 (LTB4)Cayman ChemicalCat. #: 20,110
Chemical compound, drugLeukotriene D4 (LTD4)Cayman ChemicalCat. #: 20,310
Chemical compound, drugLitocholic acidSigma-AldrichCat. #: L6250
Chemical compound, drugMelatoninBachemCat. #: Q-1300
Chemical compound, drugMDL 29,951Cayman ChemicalCat. #: 16,266
Chemical compound, drugNeuropeptide FF (NPFF)TocrisCat. #: 3,137
Chemical compound, drugNeuropeptide Y (NPY)BachemCat. #: H-6375
Chemical compound, drugNicotinic acidAbcamCat. #: ab120145
Chemical compound, drugNociceptinTocrisCat. #: 910
Chemical compound, drugNoradrenalineTocrisCat. #: 5,169
Chemical compound, drugOleoyl-Lysophosphatidic acid (O-LPA)Sigma-AldrichCat. #: L7260
Chemical compound, drugOrexin-ABachemCat. #: H-4172
Chemical compound, drugOxytocinTocrisCat. #: 1910
Chemical compound, drugParathyroid Hormone (Aittaleb et al., 2010; Aittaleb et al., 2011; Armando et al., 2014; Atwood et al., 2011; Avet et al., 2020; Azzi et al., 2003; Bowes et al., 2012; Brabet et al., 1998; Breton et al., 2010; Bünemann et al., 2003; Carr et al., 2014; Casey et al., 1990; Chandan et al., 2021; De Haan and Hirst, 2004; Devost et al., 2017; Fukuhara et al., 2001; Galandrin et al., 2007; Galés et al., 2005; Galés et al., 2006; Goupil et al., 2015; Hauser et al., 2017; Hauser et al., 2022; Hoffmann et al., 2005; Inoue et al., 2019; Jordan et al., 1999; Kawamata et al., 2003; Kenakin, 2019; Kim et al., 2002; Kobayashi et al., 2019; Laschet et al., 2019; Lavoie et al., 2002; Leduc et al., 2009; Leguay et al., 2021)Sigma-AldrichCat. #: P3796
Chemical compound, drugPertussis toxin (PTX) from Bordetella pertussisList Biological LaboratoriesCat. #: 179 A(LB)
Chemical compound, drugpH (proton) (Hydrochloric acid)Sigma-AldrichCat. #: 320,331
Chemical compound, drugProbenecidSigma-AldrichCat. #: P8761
Chemical compound, drugProlume Purple (methoxy e-Coelenterazine; Me-O-e-CTZ)NanolightCat. #: 369
Chemical compound, drugPropionate (sodium salt)Sigma-AldrichCat. #: P1880
Chemical compound, drugProstaglandin D2 (PGD2)Cayman ChemicalCat. #: 12,010
Chemical compound, drugProstaglandin E2 (PGE2)Sigma-AldrichCat. #: P0409
Chemical compound, drugRFamide-related peptide 3 (RFRP3)TocrisCat. #: 4,683
Chemical compound, drugRimonabantCayman ChemicalCat. #: 9000484
Chemical compound, drugSaralasinApexBioCat. #: B5063
Chemical compound, drugSerotoninCayman ChemicalCat. #: 14,332
Chemical compound, drugSLIGKV-NH2 (PAR2 AP)TocrisCat. #: 3,010
Chemical compound, drugSNC80Sigma-AldrichCat. #: S2812
Chemical compound, drugSomatostatin-14BachemCat. #: H-6276
Chemical compound, drugSphingosine 1-phosphateCaymanCat. #: 62,570
Chemical compound, drugTFLLR-NH2 (PAR1 AP)TocrisCat. #: 1,464
Chemical compound, drugTRV027Provided by Bristol-Myers SquibbN/A
Chemical compound, drugUBO-QIC (FR900359)Institute for Pharmaceutical Biology of the University of BonnN/A
Chemical compound, drugUndecanoic acidSigma-AldrichCat. #: 171,476
Chemical compound, drugUrocortin IIPhoenix PharmaceuticalCat. #: 019–30
Chemical compound, drugUTPSigma-AldrichCat. #: U1006
Chemical compound, drugVasoactive Intestinal Peptide (VIP)TocrisCat. #: 1911
Chemical compound, drugWB4101TocrisCat. #: 946
Chemical compound, drugWIN55,212–2Enzo Life SciencesCat. #: BMLCR105
Chemical compound, drugYM-254890Wako Pure Chemical Industries (Fujifilm)Cat. #: 257–00631
Chemical compound, drugZinc chloride (Zn2+)Sigma-AldrichCat. #: 229,997
Software, algorithmPrism, Version 8.3GraphPad
Software, algorithmMATLAB, Version R2019bMathWorks
Software, algorithmImageJ, Version 1.52 aNIH https://imagej.nih.gov/ij/
Software, algorithmScipy, Version 1.4.1https://www.scipy.org
Other96 W white plateGreiner Bio-oneCat. #: 655,083
Other96 W black plate, clear-bottomGreiner Bio-oneCat. #: 655,090
OtherOptiPlate-384, White Opaque 384-well MicroplatePerkin ElmerCat. #: 6007290
Other35 mm poly-d-lysine-coated glass-bottom culture dishesMattekCat. #: P35GC-1.5–14 C
OtherMicroplate washerBioTek InstrumentsCat. #: 405TSUS
OtherD300e Digital DispenserTecan
OtherT8 + Dispensehead CassettesHp (Tecan)Cat. #: 30097370
OtherSynergy NEO Luminescence microplate readerBioTek Instruments
OtherFlexStation 2 Multi-mode, auto-pipetting microplate readerMolecular Devices
OtherInverted microscopeNikon Eclipse Ti-U
Otherx60 objective lensNikon CFI Apochromat TIRF
OtherEMCCD cameraNuvu HNu 512

Additional files

Supplementary file 1

Impact of G proteins, receptor or effector-RlucII titration on absolute pEC50 values.

(A) Absolute pEC50 values of responses elicited in WT vs. Knockout Gα protein background cells. pEC50 values deduced from dose-response curves for various receptor tested in parental (WT) HEK293 cells or devoid of Gs, G12/13, Gq/11 or Gi/o proteins are related to Figure 2—figure supplement 1. (B) Absolute pEC50 values of responses elicited in cells transfected with different amounts of Gα proteins. pEC50 values deduced from dose-response curves obtained following Gα subunit titration in HEK293 cells transfected with GEMTA sensors and related to Figure 2—figure supplement 3A. (C) Absolute pEC50 values of responses elicited in cells transfected with different amounts of receptors. pEC50 values deduced from dose-response curves obtained following ETA titration in HEK293 cells transfected with GEMTA sensors and related to Figure 2—figure supplement 3B. (D) Absolute pEC50 values of responses elicited in cells transfected with different amounts of Effector-RlucII. pEC50 values deduced from dose-response curves obtained following Effector-RlucII titration in HEK293 cells transfected with GEMTA sensors and related to Figure 2—figure supplement 3C.

https://cdn.elifesciences.org/articles/74101/elife-74101-supp1-v2.xlsx
Supplementary file 2

List of tested receptors and ligands, along with the raw Emax, absolute pEC50 and their corresponding double normalized (dnor) values.

The Emax (in % of vehicle response) and absolute pEC50 values deduced from concentration-response curves for the 100 GPCRs tested as well as the double normalized Emax and pEC50 values calculated are related to Supplementary file 3 and Figure 3, respectively.

https://cdn.elifesciences.org/articles/74101/elife-74101-supp2-v2.xlsx
Supplementary file 3

Signaling profiles of 100 therapeutically-relevant human GPCRs using the EMTA ebBRET platform.

Concentration-response curves in HEK293 cells expressing the indicated biosensor after stimulation of heterologously expressed receptor with the indicated ligand. Data are the mean ± SEM from at least 3 independent experiments and expressed in % of the response obtained for cells treated with vehicle. For ligands that elicited endogenous receptor-mediated responses (curves with light gray and yellow background for responses similar to and better responses than those obtained with the endogenous receptors, respectively), curves from cells expressing endogenous or heterologously expressed receptors are shown in Figure 3—figure supplement 2.

https://cdn.elifesciences.org/articles/74101/elife-74101-supp3-v2.pdf
Supplementary file 4

Comparison of G protein couplings identified with EMTA platform and other datasets.

Comparison of G protein couplings identified with EMTA platform and TGF-α shedding assay in Inoue et al., 2019 (A) or reported in GtP database (B).

https://cdn.elifesciences.org/articles/74101/elife-74101-supp4-v2.xlsx
Supplementary file 5

EMTA biosensor amino acid sequences.

https://cdn.elifesciences.org/articles/74101/elife-74101-supp5-v2.docx
Transparent reporting form
https://cdn.elifesciences.org/articles/74101/elife-74101-transrepform1-v2.docx

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Charlotte Avet
  2. Arturo Mancini
  3. Billy Breton
  4. Christian Le Gouill
  5. Alexander S Hauser
  6. Claire Normand
  7. Hiroyuki Kobayashi
  8. Florence Gross
  9. Mireille Hogue
  10. Viktoriya Lukasheva
  11. Stéphane St-Onge
  12. Marilyn Carrier
  13. Madeleine Héroux
  14. Sandra Morissette
  15. Eric B Fauman
  16. Jean-Philippe Fortin
  17. Stephan Schann
  18. Xavier Leroy
  19. David E Gloriam
  20. Michel Bouvier
(2022)
Effector membrane translocation biosensors reveal G protein and βarrestin coupling profiles of 100 therapeutically relevant GPCRs
eLife 11:e74101.
https://doi.org/10.7554/eLife.74101