Preferential Gs protein coupling of the galanin Gal1 receptor in the µ-opioid-Gal1 receptor heterotetramer

Abstract

Recent studies have proposed that heteromers of µ-opioid receptors (MORs) and galanin Gal₁ receptors (Gal₁Rs) localized in the mesencephalon mediate the dopaminergic effects of opioids. The present study reports converging evidence, using a peptide-interfering approach combined with biophysical and biochemical techniques, including total internal reflection fluorescence microscopy, for a predominant homodimeric structure of MOR and Gal₁R when expressed individually, and for their preference to form functional heterotetramers when co-expressed. Results show that a heteromerization-dependent change in the Gal₁R homodimeric interface leads to a switch in G-protein coupling from inhibitory Gi to stimulatory Gs proteins. The MOR-Gal₁R heterotetramer, which is thus bound to Gs via the Gal₁R homodimer and Gi via the MOR homodimer, provides the framework for a canonical Gs-Gi antagonist interaction at the adenylyl cyclase level. These novel results shed light on the intense debate about the oligomeric quaternary structure of G protein-coupled receptors, their predilection for heteromer formation, and the resulting functional significance.

Keywords: DAMGO (PubChem CID: 5462471); Endomorphin-1 (PubChem CID: 5311080); Fentanyl (PubChem CID: 3345); G protein coupled receptor oligomerization; Galanin (PubChem CID: 16133823); Galanin receptors; M40 (PubChem CID: 16133821); M617 (PubChem CID: 16158157); Methadone (PubChem CID: 4095); Naloxone (PubChem CID: 5284596); Opioid receptors; Total internal reflection fluorescence microscopy.

Fig. 1.

Negative cooperativity of endomorphin-1 in radioligand binding experiments. Representative competitive inhibition curves of [³H]naloxone/endomorphin-1 in HEK-293T cell lines stably expressing either MOR alone (MU cells; a) or with Gal₁R (MU-GAL cells; b); data was adjusted with the ‘dimer-receptor model’, which provided a significantly better biphasic versus monophasic fit for both cell lines (p<0.05 in all cases; see Materials and Methods), with two affinities (K_DB1 and K_DB2; n = 7-9, with triplicates) and negative cooperativity (D_CB ≈ −1), indicative of a predominant population of MOR homodimers.

Fig. 2.

TM interfaces of MOR and Gal₁R homomers and heteromers in BiFC experiments. Results from BiFC experiments in HEK-293T cells co-transfected with MOR-nYFP and MOR-cYFP (a), Gal₁R-nYFP and Gal₁R-cYFP (b) or MOR-nYFP and Gal₁R-cYFP (c), in the absence (−) or the presence of the indicated TM peptides (at 4 μM; numbered 1 to 7) from MOR (green symbols and plots) or Gal₁R (blue symbols and plots), and in the absence or the presence of the co-transfected non-fused Gal₁R or MOR (lower graphs in a and b, respectively); fluorescence values (in means ± S.D.) are expressed as the percentage of the fluorescence in the absence (−) of the indicated TM peptides (n = 6, with triplicates); * represent significantly lower values as compared to control values (p < 0.001; one-way ANOVA followed by Dunnett’s multiple comparison tests). The schemes in a and b illustrate the corresponding interfaces of the MOR-MOR (a) and Gal₁R-Gal₁R (b) homomers in the absence (upper) and presence (lower) of Gal₁R and MOR, respectively. The scheme in c illustrates the computational model of the MOR-Gal₁R heterotetramer built using the experimental interfaces predicted in panels a–c (TM 5/6 for heterodimerization and TM 4/5 for homodimerization; see text).

Fig. 3.

Heterotetrameric structure of the MOR-Gal₁R heteromer in TIRF experiments. a. In the left panel, representative TIRF image showing fluorescent particles formed by mT-Gal₁R in HEK-293T cells; in the middle and right panels, example of a single mT-Gal1R particle with the time course for the change in fluorescence intensity (arbitrary units) showing two-step photobleaching. b-c. Results summarizing the number of bleaching steps for mT-Gal₁R or mC-MOR expressed separately and studied in the absence (C, control) or presence of the indicated MOR TM peptides (at 1 μM). d. In the left panel, representative TIRF image showing fluorescent particles formed by mT-Gal₁R (green), mC-MOR (red) separate and colocalized (yellow); in the middle and right panels, example of a single colocalized particle with the time course for two-photobleaching steps each for mT-Gal₁R and mC-MOR, indicating a heterotetramer. e. Result summarizing the composition of colocalized particles studied in the absence (C, control) or presence of the indicated MOR TM peptides (at 1 μM). f-g. Analysis of the non-colocalized mC-MOR and mT-Gal₁R particles from cells expressing both receptors studied in the absence (c, control) or presence of the indicated MOR TM peptides (at 1 μM). Data represent particle counts from 4–6 cells per condition studied.

Fig. 4.

G protein coupling of MOR, Gal₁R and the MOR-Gal₁R heteromer in BRET and CODA-RET experiments. a-c. Representative concentration-response curves of methadone (dark green) and fentanyl (light green) of BRET experiments from HEK-293T cells co-transfected with MOR-RLuc and the α subunit of Gi (a), Gq (b) or Gs (c) fused to YFP. e-g. Representative concentration-response curves of M617 (dark blue) and M40 (light blue) of BRET experiments from HEK-293T cells co-transfected with Gal₁R-RLuc and the α subunit of Gi (e), Gq (f) or Gs (g) fused to YFP. i-k. Representative concentration-response curves of methadone (dark green), fentanyl (light green) and M617 (dark blue) of CODA-RET experiments from HEK-293T cells co-transfected with MOR-nRLuc, Gal₁R-cRLuc and the α subunit of Gi (i), Gq (j) or Gs (k) fused to YFP. d and h. EC₅₀ values of the BRET experiments with MOR-RLuc (d) or Gal1R-RLuc (h) (in means ± S.E.M.; n = 6 in all experiments, with triplicates). l. EC₅₀ values of the CODA-RET experiments with MOR-nRLuc and Gal₁R-cRLuc (n = 5–6, with triplicates). The EC₅₀ values of fentanyl and methadone were significantly different in the CODA-RET (*: p < 0.01; two-tailed paired t test), but not in the BRET experiments.

Fig. 5.

Preferential Gs protein coupling of Gal₁R when co-expressed with MOR in cAMP formation experiments. Formation of cAMP in HEK-293T cells transfected with Gal₁R-YFP (a) MOR-RLuc (b) or both (c-e) upon exposure of forskolin (FK; 0.5 μM), galanin ligands (galanin, M617 or M40; all at 0.5 μM; blue bars within dashed frames) and endomorphin-1 (0.5 μM; green bars within dotted frame) alone or combined, in the absence (a-c) and presence (d,e) of the indicated TM peptides (TM6 or TM7; 4 μM) from MOR. “Gs-coupled” or “no-Gs-coupled” indicates the ability to increase or not cAMP formation when administered alone; “Gi-coupled” or “no-Gi-coupled” indicates the ability to decrease or not FK-induced cAMP formation; “canonical” indicates the ability of endomorphin-1 to counteract galanin-, M617- or M40-induced cAMP formation. Values are means ± S.D. (n = 6 with triplicates in all experiments) of the percentage of FK-induced cAMP formation and analyzed statistically with one-way ANOVA, followed by Tukey’s multiple comparison test (*: p < 0.001, compared with basal; #: p < 0.001, compared with FK; &: p < 0.001, compared with galanin, M617 or M40 when administered alone).

Fig. 6.

Dimeric interface-dependent hindering role of the hgh4 loop of the Gαs subunit in its ability to couple to Gal₁R. a. Sequence alignment of the C-terminal 100 amino acids of the sixteen Gα subunits, as described in GproteinDb [59]. The 13 extra residues of the hgh4 loop (in red), between HG and H4 helices of the Gαs subunits (see scheme on the left panel), that are absent in the other G protein subtypes, are highlighted in grey. b-d. Computational models of the Gal₁R-Gal₁R homodimer, constructed using the TM 5/6 interface (b, upper panel; c; d) or the TM 4/5 (b, lower panel) interface, in complex with Gs (b), Gi (c), and Gq (d). The encircled area shows the clash of the 13 extra residues of Gs with the G protein unbound protomer in the TM 5/6 interface. e. Representative concentration-response curves of the M617 agonist of BRET experiments from HEK-293T cells co-transfected with Gal₁R-RLuc and WT Gs fused to YFP (blue line; left panel) or a mutant Gαs (with the extra 13 residues being removed) fused to YFP (red curve; left panel); or cells co-transfected with Gal₁R-RLuc and WT Gi fused to YFP (blue curve; middle panel) or a mutant Gαi (with the extra 13 residues being added) fused to YFP (red curve; middle panel). In the right panel (e), comparison of E_max values of the BRET experiments between WT and mutant α-subunits (in means ± S.E.M.; n = 6 in all experiments, with triplicates). The E_max values between WT and mutant α-subunits were significantly different (*: p < 0.05, **: p < 0.01; two-tailed paired t test).