Research ArticleGPCR SIGNALING

Allosteric signaling through an mGlu2 and 5-HT2A heteromeric receptor complex and its potential contribution to schizophrenia

See allHide authors and affiliations

Science Signaling  12 Jan 2016:
Vol. 9, Issue 410, pp. ra5
DOI: 10.1126/scisignal.aab0467
  • Fig. 1 The ability of the mGlu2/3 receptor agonist LY379268 to stimulate Ca2+ release requires the assembly of 5-HT2A–mGlu2 heteromers in HEK 293 cells.

    (A to C) HEK 293 cells transfected with the control plasmid pcDNA3.1 or transfected with plasmids encoding mGlu2-eYFP and 5-HT2A–mCherry alone or in combination were loaded with Fura-2 and monitored for [Ca2+]i signals after sequential stimulation with 100 μM LY379268 (an mGlu2/3 receptor agonist) and 10 μM 5-HT. (A) Images captured before and after drug additions are representative of three independent experiments. (B) Time course of Ca2+ release in HEK 293 cells expressing the indicated constructs. The arrowheads indicate the times when the drugs were added. (C) Analysis of the fold increase in [Ca2+]i in cells expressing the indicated combination of receptors. Data in (B) and (C) are means ± SEM of five to eight experiments. (D) HEK 293 cells transfected with pcDNA3.1 as a control or cotransfected with the indicated ratios of plasmids encoding eYFP- and mCherry-tagged receptors were sequentially stimulated with LY379268 and 5-HT before Ca2+ release was measured. Data are means ± SEM of three experiments. (E) HEK 293 cells cotransfected with plasmids encoding mGlu2-eYFP and 5-HT2A–mCherry were stimulated with the indicated concentrations of LY379268 (left panel) and then with 5-HT (right panel) before Ca2+ release was measured. Data are means ± SEM of 3 to 10 experiments. (F) HEK 293 cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry were treated with vehicle (−) or were sequentially stimulated with LY379268 and 5-HT in the absence or presence of 10 μM LY341495 (an mGlu2/3 receptor antagonist). Data are means ± SEM of three to five experiments. (G) HEK 293 cells transfected with control plasmid or cotransfected with plasmids encoding the indicated eYFP- and mCherry-tagged constructs were sequentially stimulated with LY379268 (white bars) and 5-HT (black bars) before Ca2+ release was measured. Data are means ± SEM of three to six experiments. *P < 0.05, **P < 0.01, and ***P < 0.001 by Bonferroni’s post hoc test of one-way analysis of variance (ANOVA) (C, D, G, and H) or by Student’s t test (F). (H) LY379268-stimulated [35S]GTPγS binding in plasma membrane preparations of HEK 293 cells transfected with plasmids encoding the indicated eYFP- or mCherry-tagged receptors. Data are means ± SEM of two or three experiments.

  • Fig. 2 Coupling to both Gi/o and Gq/11 proteins is necessary for LY379268-induced Ca2+ release in HEK 293 cells expressing 5-HT2A–mGlu2 heteromers.

    (A) Effect of 20-min pretreatment with 10 μM M100,907 on Ca2+ release in HEK 293 cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry after sequential stimulation with 100 μM LY379268 (white bars) and 10 μM 5-HT (black bars). (B) Measurement of Ca2+ release in HEK 293 cells transfected with control plasmid or coexpressing mGlu2-eYFP and either 5-HT2A–mCherry or 5-HT2A–I181D–mCherry after sequential stimulation with LY379268 and 5-HT. (C) Effect of 20-min pretreatment with 1 μM UBO-QIC (a selective Gq/11 inhibitor) or vehicle on Ca2+ release in HEK 293 cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry after sequential stimulation with LY379268 and 5-HT. (D) Effect of 20-min pretreatment with 10 μM U73122 (a selective PLC-γ inhibitor) or vehicle on Ca2+ release in HEK 293 cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry after sequential stimulation with LY379268 and 5-HT. (E) Measurement of Ca2+ release in HEK293 cells transfected with control plasmid or coexpressing 5-HT2A–mCherry with either mGlu2-eYFP or mGlu2-F756S–eYFP after sequential stimulation with LY379268 and 5-HT. (F) Effect of overnight incubation with PTX (500 ng/ml) or vehicle on Ca2+ release in HEK 293 cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry after sequential stimulation with LY379268 and 5-HT. (G) Effect of 30-min pretreatment with 10 μM NF023 (a selective Gαi/o inhibitor) or vehicle on Ca2+ release in HEK 293 cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry after sequential stimulation with LY379268 and 5-HT. (H) Effect of 60-min pretreatment with 10 μM MPS (a selective Gβγ blocker) or vehicle on Ca2+ release in cells coexpressing mGlu2-eYFP and 5-HT2A–mCherry after sequential stimulation with LY379268 and 5-HT. (I and J) Effect of 10 μM MPS on G protein–coupled inwardly rectifying potassium (GIRK) currents in response to 100 μM GTPγS included in the patch pipette. Representative traces (I) and the summary of data (J) from whole-cell patch-clamp recordings show the membrane-permeable sequence (MPS)–dependent inhibition of GTPγS-induced GIRK currents in HEK 293 cells expressing GIRK1 and GIRK4. Barium (Ba) inhibited GIRK currents and enabled the subtraction of GIRK-independent currents. (K) Measurement of Ca2+ release in response to sequential stimulation with LY379268 and 5-HT in HEK 293 cells transfected with control plasmid or cotransfected with plasmids encoding the indicated eYFP- and mCherry-tagged constructs or with plasmids encoding the indicated mGlu2 mutant. Data are means ± SEM of 3 or 4 experiments (A, C, D, G, and H), 3 to 5 experiments (E), 4 or 5 experiments (F), 3 to 6 experiments (B and K), or 10 experiments (J). *P < 0.05, **P < 0.01, and ***P < 0.001 by Bonferroni’s post hoc test of one-way ANOVA (A to H and K) or Student’s t test (J). (L) LY379268-stimulated [35S]GTPγS binding in plasma membrane preparations of HEK 293 cells expressing the indicated eYFP- or mCherry-tagged receptors or coexpressing the indicated mCi-N172– or mCi-C67–tagged constructs. Data are means ± SEM of two or three experiments each performed in triplicate. n.s., not significant.

  • Fig. 3 The relative orientation of the two mGlu2 protomers affects 5-HT2A receptor–dependent Gq/11 coupling.

    (A and B) Cartoon representation of the relative positions of the components of the 5-HT2A–mGlu2 heteromeric receptor complex that are necessary to induce Ca2+ release in the presence of the mGlu2/3 agonist LY379268 based on the mGlu2/mGlu3 chimeric and single point mutation constructs expressed in HEK 293 cells. (C) Measurement of Ca2+ release in HEK 293 cells that were left untransfected or were cotransfected with plasmids encoding the indicated constructs after sequential stimulation with LY379268 and 5-HT. Data are means ± SEM of three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001 by Bonferroni’s post hoc test of one-way ANOVA. (D) Measurement of LY379268-stimulated [35S]GTPγS binding in plasma membrane preparations of HEK 293 cells coexpressing mCi-N172– and mCi-C67–tagged receptors. Data are means ± SEM of four to eight experiments, each performed in triplicate.

  • Fig. 4 Single-cell imaging and structure of oligomers of the 5-HT2A–mGlu2 heteromeric receptor complex.

    (A to C) Top: Cartoons illustrate the basis of direct and sequential FRET. Bottom: Three-color FRET (3-FRET) signals in cells expressing c-Myc–5-HT2A–Cerulean, HA-mGlu2-mCherry, and either c-Myc–5-HT2A–mCitrine or c-Myc–5-HT2A–Y67C-mCitrine. (A) Sequential 3-FRET was calculated by subtracting the signal in cells expressing Cerulean, mCitrine, and mCherry. Data are means ± SEM of 16 to 20 cells per group in three independent experiments. **P < 0.01 by Bonferroni’s post hoc test of one-way ANOVA. (B and C) Single FRET signals in cells expressing c-Myc–5-HT2A–Cerulean/c-Myc–5-HT2A–mCitrine and HA-mGlu2-mCherry. The single FRET signal from c-Myc–5-HT2A–Cerulean/c-Myc–5-HT2A–mCitrine (B) and the single FRET signal from c-Myc–5-HT2A–mCitrine/HA-mGlu2-mCherry (C) were used to establish the appropriate conditions to measure both direct two-protein and sequential three-protein FRET. Data are means ± SEM of 16 to 20 cells per group in three independent experiments. ***P < 0.001 by Student’s t test. (D and E) TM4 of 5-HT2A mediates complex formation with the mGlu2 receptor. Top: Schematic of the 5-HT2A/5-HT2C chimeras studied. Bottom: HEK 293 cells coexpressing HA-mGlu2 and either c-Myc–5-HT2A or c-Myc–5-HT2C (D) or coexpressing HA-mGlu2 and either c-Myc–5-HT2AΔTM1 or c-Myc–5-HT2AΔTM4 (E) were subjected to coimmunoprecipitation (IP) with antibody against the c-Myc tag and then were analyzed by Western blotting (WB) with antibodies against the indicated tags. As controls, HEK 293 cells separately expressing the c-Myc– or HA-tagged forms of the indicated receptors were mixed. Western blots are representative of three independent experiments.

  • Fig. 5 Model for the docking of Gi/o onto the 5-HT2A–mGlu2 heteromeric receptor complex.

    (A to C) Ribbon representation of the vertical (A), extracellular (B), and cytoplasmic (C) views of the mechanism by which coupling of Gi/o proteins to the mGlu2 protomer located distal to the 5-HT2A component is necessary to enable allosteric crosstalk with 5-HT2A and consequently activate Gq/11-dependent signaling. The homodimeric interfaces of mGlu2 and 5-HT2A involve residues from TM1. The interface of the 5-HT2A–mGlu2 heteromeric complex involves residues from TM4. αAH, α-helical domain of the Gα subunit; αRas, Ras-like domain of the Gα subunit.

  • Fig. 6 Activation of Gq/11 by the mGlu2/3 agonist LY379268 is reduced in the frontal cortex of 5-HT2A knockout mice.

    (A and B) Validation of the subcellular fractionation protocol performed to isolate PSDs and PAZs from the mouse frontal cortex. Equal amounts of proteins from mouse frontal cortex PSD and PAZ fractions were resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed by Western blotting with antibodies against postsynaptic (PSD-95) and presynaptic (syntaxin-1) markers (A), as well as with antibodies specific for mGlu2 and 5-HT2A (B). Western blots are representative of three independent experiments. (C and D) [35S]GTPγS binding to frontal cortex membrane preparations of wild-type and mGlu2 knockout mice after treatment with vehicle or 10 μM LY379268 in the absence or presence of 10 μM LY341495 (the mGlu2/3 receptor antagonist) followed by immunoprecipitation with anti-Gαi1,2,3 (C) or anti-Gαq/11 (D) antibodies. Data are means ± SEM of three or four independent experiments, performed in triplicate or quintuplicate for each condition. (E and F) [35S]GTPγS binding to frontal cortex membrane preparations of wild-type and 5-HT2A knockout mice after treatment with vehicle or 10 μM LY379268 in the absence or presence of 10 μM LY341495 followed by immunoprecipitation with anti-Gαi1,2,3 (E) or anti-Gαq/11 (F) antibodies. Data are means ± SEM of three or four independent experiments, performed in triplicate or quintuplicate for each condition. *P < 0.05, **P < 0.01, and ***P < 0.001 by Bonferroni’s post hoc test of two-way ANOVA.

  • Fig. 7 Activation of Gq/11 by the mGlu2/3 agonist LY379268 is dysregulated in the postmortem brains of schizophrenic subjects.

    (A and B) [35S]GTPγS binding in postmortem frontal cortexes of schizophrenic subjects and normal controls after treatment with vehicle or the indicated concentrations of LY379268 in the absence or presence of 10 μM LY341495 (the mGlu2/3 receptor antagonist) followed by immunoprecipitation with anti-Gαi1,2,3 (A) or anti-Gαq/11 (B) antibodies. Data are means ± SEM of 27 independent experiments, performed in triplicate or quintuplicate for each condition. See table S1 for demographic information and table S2 for pharmacological data. **P < 0.01 and ***P < 0.001 by Bonferroni’s post hoc test of two-way ANOVA.

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/9/410/ra5/DC1

    Fig. S1. Effect of ATP on Ca2+ release.

    Fig. S2. Relative abundances of eYFP- and mCherry-tagged constructs in HEK 293 cells.

    Fig. S3. Concentration-response curves of LY404039 and l-glutamate in HEK 293 cells.

    Fig. S4. LY379268 has no effect on Ca2+ release in cells coexpressing 5-HT2C and mGlu2 receptors.

    Fig. S5. Characterization of 5-HT2A, 5-HT2C, mGlu2, and mGlu3 constructs.

    Fig. S6. Measurement of molecular interactions between eYFP- and mCherry-tagged constructs by FCM-based FRET assay.

    Fig. S7. Radioligand binding assays.

    Fig. S8. Characterization of 3-FRET constructs.

    Fig. S9. Western blotting in frontal cortex of wild-type and 5-HT2A knockout mice.

    Table S1. Demographic characteristics of schizophrenic subjects and controls.

    Table S2. [35S]GTPγS binding in postmortem human brain samples.

  • Supplementary Materials for:

    Allosteric signaling through an mGlu2 and 5-HT2A heteromeric receptor complex and its potential contribution to schizophrenia

    José L. Moreno, Patricia Miranda-Azpiazu, Aintzane García-Bea, Jason Younkin, Meng Cui, Alexey Kozlenkov, Ariel Ben-Ezra, Georgios Voloudakis, Amanda K. Fakira, Lia Baki, Yongchao Ge, Anastasios Georgakopoulos, José A. Morón, Graeme Milligan, Juan F. López-Giménez, Nikolaos K. Robakis, Diomedes E. Logothetis, J. Javier Meana,* Javier González-Maeso*

    *Corresponding author. E-mail: javier.meana{at}ehu.eus (J.J.M.); jgmaeso{at}vcu.edu (J.G.-M.)

    This PDF file includes:

    • Fig. S1. Effect of ATP on Ca2+ release.
    • Fig. S2. Relative abundances of eYFP- and mCherry-tagged constructs in HEK 293 cells.
    • Fig. S3. Concentration-response curves of LY404039 and ʟ-glutamate in HEK 293 cells.
    • Fig. S4. LY379268 has no effect on Ca2+ release in cells coexpressing 5-HT2C and mGlu2 receptors.
    • Fig. S5. Characterization of 5-HT2A, 5-HT2C, mGlu2, and mGlu3 constructs.
    • Fig. S6. Measurement of molecular interactions between eYFP- and mCherry-tagged constructs by FCM-based FRET assay.
    • Fig. S7. Radioligand binding assays.
    • Fig. S8. Characterization of 3-FRET constructs.
    • Fig. S9. Western blotting in frontal cortex of wild-type and 5-HT2A knockout mice.
    • Table S1. Demographic characteristics of schizophrenic subjects and controls.
    • Table S2. [35S]GTPγS binding in postmortem human brain samples.

    [Download PDF]

    Technical Details

    Format: Adobe Acrobat PDF

    Size: 3.92 MB


    Citation: J. L. Moreno, P. Miranda-Azpiazu, A. García-Bea, J. Younkin, M. Cui, A. Kozlenkov, A. Ben-Ezra, G. Voloudakis, A. K. Fakira, L. Baki, Y. Ge, A. Georgakopoulos, J. A. Morón, G. Milligan, J. F. López-Giménez, N. K. Robakis, D. E. Logothetis, J. J. Meana, J. González-Maeso, Allosteric signaling through an mGlu2 and 5-HT2A heteromeric receptor complex and its potential contribution to schizophrenia. Sci. Signal. 9, ra5 (2016).

    © 2016 American Association for the Advancement of Science

Stay Connected to Science Signaling

Navigate This Article