o activation by rat melanin-concentrating hormone receptor 1

Cellular Signalling 27 (2015) 818–827

Contents lists available at ScienceDirect

Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig

Identification of amino acids that are selectively involved in Gi/o activation by rat melanin-concentrating hormone receptor 1 Akie Hamamoto, Yuki Kobayashi, Yumiko Saito ⁎ Graduate School of Integrated Arts and Sciences, Hiroshima University, Hiroshima, Japan

a r t i c l e

i n f o

Article history: Received 9 November 2014 Received in revised form 30 December 2014 Accepted 14 January 2015 Available online 22 January 2015 Keywords: GPCR Melanin-concentrating hormone (MCH) Melanin-concentrating hormone receptor 1 (MCHR1) Mutation Signal transduction G-protein

a b s t r a c t Many G-protein-coupled receptors (GPCRs) are known to functionally couple to multiple G-protein subfamily members. Although promiscuous G-protein coupling enables GPCRs to mediate diverse signals, only a few GPCRs have been identified with differential determinants for coupling to distinct Gα proteins. Mammalian melanin-concentrating hormone receptor 1 (MCHR1) couples to dual G-protein subfamilies. However, the selectivity mechanisms between MCHR1 and different subtypes of Gα proteins are unclear. Our previous studies demonstrated that mammalian MCHR1 couples to both Gi/o and Gq, whereas goldfish MCHR1 exclusively couples to Gq. In this study, we analyzed multiple sequence alignments between rat and goldfish MCHR1s, and designed three multisubstituted mutants of rat MCHR1 by replacing corresponding residues with those in goldfish MCHR1, focusing on regions around the cytosolic intracellular loops. By measurement of intracellular Ca2+ mobilization, we found that two MCHR1 mutants, i2_6sub and i3_6sub, which contained six simultaneously substituted residues in the second intracellular loop or a combination of substituted residues in the third intracellular loop and fifth transmembrane domain, respectively, significantly reduced Gi/o-sensitive pertussis toxin responsiveness without altering Gq-mediated activity. Analyses of 10 other substitutions revealed that the multiple substitutions in i2_6sub and i3_6sub were necessary for Gi/o-selective responses. As judged by Gi/odependent GTPγS binding and cyclic AMP assays, i2_6sub and i3_6sub elicited phenotypes for impaired Gi/omediated signaling. We also monitored the dynamic mass redistribution (DMR) in living cells, which reveals receptor activity as an optical trace containing activation of all GPCR coupling classes. Cells transfected with i2_6sub or i3_6sub exhibited reduced Gi/o-mediated DMR responses compared with those transfected with MCHR1. These data suggest that two different regions independently affect the Gi/o-protein preference, and that multiple residues comprise a conformation favoring Gi/o-protein coupling and subsequently result in Gi/o-selective signaling. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Melanin-concentrating hormone (MCH) is a cyclic neuropeptide composed of 19 amino acids in mammals. MCH-expressing neurons are predominantly localized in the lateral hypothalamus, which is known as the center for feeding behavior and energy expenditure [1–3]. The effects of MCH are mediated through two class A G-proteincoupled receptors (GPCRs), MCHR1 and MCHR2 [4–6], of which MCHR2 is not functionally present in rodents [7]. MCHR1 is expressed at high levels in several brain regions [8], and MCHR1-knockout mice Abbreviations: cAMP, cyclic AMP; DMR, dynamic mass redistribution; GPCR, G-proteincoupled receptor; GTPγS, guanosine 5′-O-[gamma-thio]triphosphate; i1, first intracellular; i2, second intracellular; i3, third intracellular; MCH, melanin-concentrating hormone; MCHR, melanin-concentrating hormone receptor; PTX, pertussis toxin; TM, transmembrane domain. ⁎ Corresponding author at: Graduate School of Integrated Arts and Sciences, Hiroshima University, 1-7-1 Kagamiyama, Higashi-hiroshima, Hiroshima 739-8521, Japan. Tel.: +81 82 424 6563; fax: +81 82 424 0759. E-mail address: [email protected] (Y. Saito).

http://dx.doi.org/10.1016/j.cellsig.2015.01.008 0898-6568/© 2015 Elsevier Inc. All rights reserved.

show decreased body weight and increased activity and energy metabolism [9,10]. Rodent behavior studies also indicated that selective MCHR1 antagonists decrease food intake and body weight [11,12]. Furthermore, several antagonists exhibit antidepressant and anxiolytic effects [13,14]. Therefore, the MCH–MCHR1 system could be an important target for the treatment of obesity, anxiety, and depression in mammals. Beside mammals, MCH and MCHRs exist in other vertebrates [15]. MCH not only causes aggregation of melanophores in teleost skin, but regulates feeding behavior via MCHR1 and/or MCHR2 in goldfish [16, 17]. Recently, we cloned and characterized four MCHR subtypes from the amphibian Xenopus tropicalis and indicated the involvement of two MCHRs in melanin-granule-concentrating and -dispersing activities [18]. Many GPCRs can activate more than one G-protein subfamily member. In mammalian cell expression systems, rat MCHR1 promiscuously couples to both Gi/o and Gq, resulting in activation of multiple signaling pathways including Ca2+ mobilization, phosphorylation of extracellular signal-regulated kinase, and inhibition of cyclic AMP (cAMP) generation [4,5,19]. Extensive mutagenesis analyses have identified several amino

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

acid residues that are essential for mammalian MCHR1 activation. The highly conserved DRY motif in MCHR1 has a role in governing receptor conformation and dual G-protein coupling/recognition [20]. Another study indicated the importance of complete glycosylation of MCHR1 (Asn13, Asn16, and Asn23) for efficient trafficking [21]. Molecular modeling of MCHR1 with MCH demonstrated that Asp123 in the third transmembrane domain (TM3) of MCHR1 is crucial for ligand binding [22]. In the second intracellular (i2) loop, Arg155 was found to be a key residue for receptor signaling via dual G-protein-mediated pathways [23]. In addition, Thr255, which is located at the junction of the third intracellular (i3) loop and sixth transmembrane domain (TM6), is important for receptor folding and correct trafficking to the cell surface [24]. In the intracellular C-terminus, two dibasic amino acids (Arg319 and Lys320) in helix 8, a common short amphiphilic helical domain, are essential for Gi/o- and Gq-mediated signaling [25]. Another mutant in the C-terminus, T317A/S325A/T342A, has no effects on the signal transduction for Ca2 + mobilization, but significantly prevents MCH-induced receptor internalization through protein kinase C and β-arrestin 2-dependent processes [26,27]. To date, there remain fundamental questions about how the G-protein selectivity of GPCRs is determined. We previously reported that F318K in the highly conserved NPxxY(x)5,6F motif in MCHR1 provides an efficient signaling property that selectively increases the Gq-mediated pathway [28]. However, the distinct amino acid residues required for optimal Gi/o-protein responses in mammalian MCHR1 have not been identified. Most GPCR studies have emphasized that membrane-proximal regions in the i2 and i3 loops and/or the C-terminal tail of the receptor play prominent roles in coupling to G-proteins [29–31]. However, only a few GPCRs belonging to class A have been demonstrated to contain molecular determinants for coupling to distinct G-proteins [32–34]. Using mammalian cell-based assays, we found that the signaling properties of MCHR1 are quite different between mammals and fish. Thus, diverse transduction assays revealed that rat MCHR1 promiscuously couples to both Gi/o and Gq, while goldfish MCHR1 exclusively couples to Gq [5,35,36]. Based on these different signaling features, we wanted to identify the regions of rat MCHR1 responsible for Gi/o coupling. First, multiple sequence alignments were made between rat and goldfish MCHR1s, and then three mutants of rat MCHR1 containing fish type-substituted helix and/or intracellular cytoplasmic loop residues were constructed. Each mutant was tested for its capacity of Gi/o coupling by measurement of intracellular Ca2+ mobilization. Next, we examined the phenotypes induced using Gi/o-mediated guanosine 5′-O[gamma-thio]triphosphate (GTPγS) binding and cAMP assays. Finally, the MCH-induced responses in living cells were measured by dynamic mass redistribution (DMR) assays, which involve multiple G-proteinmediated pathways. We identified two mutants, in which multiple residues in the i2 loop or i3 loop/TM5 were simultaneously substituted, that caused decreases in Gi/o-mediated signaling without changing the MCHR1 activity via Gq protein. 2. Materials and methods 2.1. cDNA constructs for rat MCHR1 and mutant receptors The generation of a cDNA encoding a Flag epitope tag before the first methionine of rat MCHR1 (NM_031758/GenBank/EMBL) was described previously [21]. Wild-type MCHR1 and Flag-tagged MCHR1 (FlagMCHR1) have similar EC50 values for MCH, indicating that the addition of the Flag-tag did not affect the receptor function. Genetyx5 software (Genetyx Corporation, Tokyo, Japan) was used to process the nucleotide and amino acid sequences of rat and goldfish, and perform the amino acid sequence alignments. Substitution mutants around the intracellular regions were produced by oligonucleotide-mediated site-directed mutagenesis using a QuikChange site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA). All mutations in the MCHR1 cDNA sequence were confirmed by sequencing analysis. Mutated MCHR1

819

cDNAs were excised by digestion with EcoRI and XhoI and inserted into the pcDNA3.1/Zeo(+) expression vector. 2.2. Cell culture and transfection Human embryonic kidney HEK293T cells were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (FBS). The plasmid DNA was mixed with FuGENE HD transfection reagent (Promega, Madison, WI), diluted with Opti-MEM (Life technologies, Carlsbad, CA), and added to 70–80% confluent cells [21]. The transfected cells were replated onto LAB-TEK 8-well plates (Nunc, Rochester, NY) for immunocytochemical analyses, 96-well plates (BIOCOAT; Becton Dickinson, Belford, MA) for Ca2+ mobilization assays, and 100mm cell culture dishes for GTPγS-binding assays. For FACScan flow cytometric analyses and cAMP assays, the cells were reseeded onto 24-well plates (Becton Dickinson). The transfected cells were cultured for a further 18–24 h. For stable transfection, the transfected cells were selected in the presence of zeocin at a final concentration of 0.4 mg/ml for 3 weeks and used for measurements of the cAMP levels. 2.3. Western blotting Western blotting analyses were performed as described previously [23]. To generate whole-cell extracts, transiently transfected HEK293T cells or CHO-K1 cells were washed with phosphate-buffered saline (PBS), and lysed with ice-cold sodium dodecyl sulfate (SDS) sample buffer (50 mM Tris–HCl pH 6.8, 2% SDS, 50 mM β-mercaptoethanol, 10% glycerol). The lysates were homogenized at 4 °C by sonication (SONICAOR Ultrasonic processor W-225; Wakenyaku Ltd., Kyoto, Japan) using five 30-s bursts at 20% power. The proteins were separated by SDS-PAGE and electrotransferred to Hybond-P PVDF membranes (GE Healthcare UK Ltd., Little Chalfont, UK). After blocking with 5% skim milk, Flag-MCHR1 on the membranes was detected by incubation with 1 μg/ml anti-DYKDDDDK primary antibody (Wako, Osaka, Japan), followed by a horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibody (GE Healthcare UK Ltd.). The reactive bands were visualized with an enhanced chemiluminescence (ECL) reagent (GE Healthcare UK Ltd.) and analyzed by ImageJ software (National Institutes of Health, Bethesda, MD). 2.4. Immunofluorescence microscopy Transfected HEK293T cells were fixed in 3.7% paraformaldehydePBS solution for 15 min. After two washes with PBS, cells with or without permeabilization (0.05% Triton X-100 in PBS for 5 min) were transferred into a blocking solution (20% goat serum in PBS) for 30 min, and then incubated with 0.5 μg/ml anti-DYKDDDDK antibody at 4 °C overnight. The bound antibodies were detected using an Alexa Fluor 488conjugated goat anti-mouse IgG secondary antibody (Molecular Probes, Eugene, OR). Fluorescence imaging was performed using a FLUOVIEW FV1000 confocal microscope (Olympus, Tokyo, Japan). Confocal images were opened in ImageJ software and relative intensity was quantified. The relative intensity represents the ratio pixel density/cell, and statistical analysis of staining intensity was performed with a Student's t-test. 2.5. FACScan flow cytometric analysis of cell surface receptors FACScan flow cytometric analyses were performed as described [21]. Transfected HEK293T cells in 24-well plates were fixed in 1.5% paraformaldehyde-PBS solution for 10 min at room temperature, and then incubated with 0.67 μg/ml anti-DYKDDDDK antibody in PBS containing 20% FBS for 1 h. After three washes with PBS, the cells were incubated with the Alexa Fluor 488-conjugated goat anti-mouse IgG secondary antibody for 1 h. The cells were washed, collected from the wells with 5 mM EDTA, and analyzed using a FACSCalibur flow cytometer (Becton Dickinson Immunocytometry Systems Inc., Franklin

820

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

Lakes, NJ). Cells were gated by light scattering or exclusion of propidium iodide, and 10,000 cells were acquired for each time point. The mean fluorescence of all cells, minus the mean cell fluorescence with the Alexa Fluor 488-conjugated secondary antibody alone, was used for the calculations.

incubated with 1 μM forskolin and various concentrations of human MCH for 15 min. The reactions were terminated with 0.3 N HCl, and the levels of extracted intracellular cAMP were measured using a radioimmunoassay kit (Cyclic AMP kit “Yamasa”; Yamasa Corp., Chiba, Japan) following the manufacturer's protocol as described previously [25].

2.6. Radioligand binding

2.10. DMR assay

Radioligand binding assays were performed as described previously [26]. Transfected HEK293T cells were scraped into ice-cold PBS and centrifuged at 1000 ×g for 5 min. The cell pellets were homogenized in icecold 50 mM Tris–HCl buffer (pH 7.4) containing 5 mM EDTA and ultracentrifuged twice at 48,000 ×g for 20 min each time at 4 °C. The pellets were suspended in 50 mM Tris–HCl buffer (pH 7.4) containing 5 mM EDTA and used as the membrane fractions. Aliquots of the membrane fractions (20 μg protein/assay) were incubated with increasing concentrations of [125I] (Phe13, Tyr19) MCH (PerkinElmer, Santa Clara, CA) from 0.02–4 nM in the absence or presence of 1 μM nonlabeled human MCH (Peptide Institute, Osaka, Japan) in 300 μl of assay buffer (50 mM Tris–HCl pH 7.4, 1 μM phosphoramidon, 0.5 mM phenylmethylsulfonylfluoride, 0.2% BSA) at room temperature for 2 h. The binding reactions were terminated by rapid filtration through GF/ C glass microfiber filters (Whatman International Ltd., Maidstone, UK) presoaked in 0.2% polyethylenimine, followed by three washes with 3 ml of PBS. The radioactivities retained in the filters were determined using a γ-counter ARC-380CL (ALOKA, Tokyo, Japan). Specific binding was defined as the difference between total binding and nonspecific binding.

DMR assays in microplates were performed using an EnSpire labelfree system (PerkinElmer) following the manufacturer's protocol (http://www.perkinelmer.com/pdfs/downloads/TCH_Comparison_of_ Performance_Enspire_Multimode_Plate_Reader_and_Corning_Epic_ System.pdf). Transiently transfected Chinese hamster ovary (CHO)-K1 cells (25,000 cells/well) were cultured for 16–18 h in EnSpire-LFC 96well cellular assay microplates-uncoated (PerkinElmer). The medium was then exchanged for assay buffer (HBSS with 20 mM HEPES, pH 7.5) and the cells were incubated inside the EnSpire label-free system for 2 h at room temperature. After a steady baseline was established, the ligand solutions were added to individual wells and the DMR was monitored for 1 h. The incubation times for pretreatment with 200 ng/ml PTX were 16–20 h. Quantification of DMR was performed by the real response at a specific time point.

2.7. Measurement of intracellular Ca2+ Measurements of intracellular Ca2+ were carried out as described [21]. Transiently transfected HEK293T cells seeded on 96-well plates were loaded with a non-wash calcium dye (Calcium Assay Kit 5; Molecular Devices, Sunnyvale, CA) in Hank's balanced salt solution (HBSS) containing 20 mM HEPES (pH 7.5) for 1 h at 37 °C. For each concentration of human MCH, the level of [Ca2 +]i was detected using a FlexStation 3 Microplate Reader (Molecular Devices). To evaluate the involvement of Gi/o activity in Ca2+ mobilization, the transfected cells were pretreated with 200 ng/ml pertussis toxin (PTX) for 18 h. The data were expressed as fluorescence (arbitrary units) versus time. The EC50 values for MCH were obtained from sigmoidal fits using a nonlinear curve-fitting program (Prism v3.0; GraphPad Software, San Diego, CA). 2.8. GTPγS-binding assay GTPγS-binding assays were performed as described previously [23]. Aliquots (10 μg) of the membrane fractions isolated for radioligand binding as described above were incubated in GTPγS binding buffer (20 mM HEPES-NaOH pH 7.5, 100 mM NaCl, 5 mM MgCl2, 0.2% BSA, 3 μM GDP) containing 0.2 nM [35S]GTPγS (PerkinElmer) and various concentrations of human MCH for 30 min at 30 °C. To determine the nonspecific binding, unlabeled GTPγS was added to the binding mixtures to a final concentration of 100 μM. The bound [35S]GTPγS was separated from free [35S]GTPγS by rapid filtration through GF/C filters and washed three times with 3 ml of ice-cold binding buffer. The radioactivities of the filters were counted in 8 ml of a scintillation cocktail (Emulsion-Scintillator Plus; Packard Bioscience, Groningen, The Netherlands) using a liquid scintillation counter LSC-6100 (ALOKA). 2.9. Measurement of cAMP production Stably transfected HEK293T cells seeded on 24-well plates were preincubated with cAMP assay buffer (HBSS with 20 mM HEPES and 0.3 mM 3-isobutyl-1-methyl-xanthine, pH 7.5) for 10 min, and then

3. Results 3.1. Effects of substitution mutations on receptor expression and ligand binding To estimate the region responsible for Gi/o responses with MCHR1, we analyzed multiple sequence alignments between rat and goldfish MCHR1s, and focused on the regions around the intracellular loops. We then mutated rat Flag-tagged MCHR1 (Flag-MCHR1) by replacing its residues with corresponding residues in goldfish MCHR1 and constructed three mutants, named i1_7sub (S69N/L71F/H72R/W73A/ C74Q/S75Q/N76T), i2_6sub (S150R/S151F/T152N/K153H/K156T/ S158C), and i3_6sub (Y228F/V229F/R234N/A242L/S243P/T257M) (Fig. 1). The i1_7sub and i2_6sub mutants contained simultaneous substitutions of seven and six residues within the first intracellular (i1) and i2 loops, respectively. The i3_6sub mutant contained a combination of mutations in two domains, that is, two residues in TM5 and four residues in the i3 loop. First, utilizing transient transfection of Flag-MCHR1 or mutant receptors into HEK293T cells, we examined the receptor expression levels and glycosylation patterns by western blotting analyses. Several immunoreactive bands were detected in the whole-cell lysates expressing Flag-MCHR1 (Fig. 2A), some of which corresponded to the predicted molecular masses of 35, 44, 45, and 60 kDa. In the cells transfected with i1_7sub or i2_6sub, the levels of mutant expression were nearly the same as those of Flag-MCHR1. However, the intensity of the 60 kDa receptor in the i3_6sub mutant was significantly reduced by 72% with that of Flag-MCHR1. Next, the cell surface expression levels of the mutants were monitored by confocal immunofluorescence microscopy. In nonpermeabilized cells, Flag-MCHR1 and two mutants (i1_7sub and i2_6sub) were clearly localized in the plasma membrane (Fig. 2B, upper). Quantification of imaging analysis showed that the i2_6sub gave an approximately 25% elevation in the cell surface expression, whereas the level of the i3_6sub was significantly reduced by nearly 35% with that of Flag-MCHR1. In cells permeabilized with Triton X100, Flag-MCHR1 and all three mutants were predominantly detected in the plasma membrane, but not in the perinuclear zone and cytoplasm (Fig. 2B, bottom). These results suggest that the mutants were efficiently trafficked to the plasma membrane. Next, the cell surface expressions were quantified by FACScan flow cytometry. The expression level of i1_7sub was equivalent to that of Flag-MCHR1, while i2_6sub showed a significant increase. In contrast, the expression level of i3_6sub was reduced by approximately 30%, compared with Flag-MCHR1 (Table 1).

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

821

Fig. 1. Targeting amino acid residues in the potential Gi/o-protein coupling domains of rat MCHR1 (nr) using goldfish MCHR1 (gf). Amino acid sequence alignments were performed between rat and teleost goldfish MCHR1s by focusing on regions around the three cytoplasmic intracellular loops (i1, i2, and i3) and transmembrane domain 5 (TM5), respectively. Three mutants, i1_7sub, i2_6sub, and i3_6sub, were designed based on rat MCHR1. The accession numbers used were as follows: Norway rat (nr), NP_113946; goldfish (gf), BAH70338. The residues located in transmembrane domains are shaded. The underlines and boldface letters show the substituted residues. TM, transmembrane domain. Three mutants in rat MCHR1 containing the fish type-substituted helix and/or intracellular cytoplasmic loop residues are shown.

A kDa

64 47 34

B

Flag-

-TX100

+TX100

Bar: 10μm Fig. 2. Determination of receptor expression by western blotting and immunofluorescence microscopy. (A) Protein expression of Flag-MCHR1 and mutant receptors evaluated by western blotting analyses. Transiently transfected HEK293T cells were lysed with SDS-sample buffer, and then homogenized at 4 °C by sonication. The proteins were separated by 12.5% SDS-PAGE, transferred to a PVDF membrane, and immunoblotted with an anti-DYKDDDDK antibody. Four major immunoreactive bands of 35, 44, 45, and 60 kDa are present in Flag-MCHR1 and individual mutant receptors, while the intensity of the 60-kDa receptor (arrow) appears to be decreased in the i3_6sub mutant. (B) Confocal immunolocalization of Flag-MCHR1 and mutant receptors. The cell surface expressions were compared using transiently transfected nonpermeabilized HEK293T cells (−TX100, without Triton X-100; upper row) and permeabilized cells HEK293T (+TX100, with Triton X-100; lower row). Flag-MCHR1 and mutants were detected using an anti-DYKDDDDK primary antibody, followed by an Alexa Fluor 488-conjugated goat anti-mouse IgG secondary antibody. Fluorescence imaging was performed using laser confocal microscopy. Vector-transfected cells incubated with the anti-DYKDDDDK antibody showed no positive staining (data not shown).

822

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827 Table 1 Cell surface expressions of Flag-MCHR1 and mutant receptors. Receptor

Cell surface expression (FACS, %)

Flag-MCHR1 i1_7sub i2_6sub i3_6sub

100 103.1 ± 2.7 120.2 ± 5.9a 72.5 ± 4.9b

The cell surface expression levels of mutants were analyzed by FACScan flow cytometry, and are shown as the mean fluorescence intensities. HEK293T cells transiently transfected with Flag-MCHR1 and mutants were stained with an anti-DYKDDDDK primary antibody, followed by staining with an Alexa Fluor 488-conjugated secondary antibody. Cells were gated by light scattering or exclusion of propidium iodide, and 10,000 cells were acquired for each time point. The data represent means ± SEM of three or four independent experiments performed in triplicate. a P b 0.05, significant difference from Flag-MCHR1 by Student's t-test. b P b 0.01, significant difference from Flag-MCHR1 by Student's t-test.

Radioligand saturation binding was analyzed by specific binding of [125I] (Phe13, Tyr19) MCH in the presence of 1 μM cold MCH. Both the Kd and maximal binding capacity (Bmax) values for the three mutants were slightly changed, but no significant differences were observed in comparison to Flag-MCHR1 (Table 2). Therefore, we confirmed that Flag-MCHR1 and the three mutants properly retained the ligandbinding capacity. 3.2. Evaluation of Gi/o- and Gq-mediated activities by measurements of intracellular Ca2 + mobilization in cells expressing MCHR1-substituted mutants MCH-induced Ca2 + release is caused via both Gi/o- and Gqmediated pathways in mammalian expression systems. To determine whether the mutant receptors had reduced Gi/o activity, the changes in [Ca2 +]i were quantified in transiently transfected HEK293T cells pretreated with Gi/o-sensitive PTX using a Flexstation [23]. PTX is known to inhibit the signaling mediated through Gi/o without altering the signaling mediated by Gq. As shown in Fig. 3, the cells expressing Flag-MCHR1 responded to MCH in a robust and dose-dependent manner. PTX pretreatment significantly increased the EC50 value for MCHinduced Ca2+ mobilization in cells expressing Flag-MCHR1, although the maximum response to MCH was not significantly affected (Fig. 3A–C). The EC50 value in PTX-treated cells was 12.12 ± 3.33 nM, while that in PTX-untreated cells was 3.77 ± 0.85 nM, giving a + PTX/− PTX ratio of 3.2 (Table 3). The + PTX/− PTX ratio for i1_7sub (2.9) was nearly equivalent to that for Flag-MCHR1 (Fig. 3A). Although we designed and analyzed five other mutants based on i1_7sub (C74Q, W73A/C74Q, W73A/C74Q/S75Q, L71F/H72R/W73A/C74Q/S75Q, and L71F/H72R/W73A/C74Q/S75Q/N76T), all of them showed equivalent ratios to that of Flag-MCHR1 (data not shown). However, in the cases of i2_6sub and i3_6sub, PTX did not induce any significant reductions in receptor activity, and the + PTX/− PTX ratios were 1.3 and 1.7, respectively (Fig. 3B, C, Table 3). Taken together, the i2_6sub and Table 2 Specific radioligand binding activities of Flag-MCHR1 and mutant receptors. Receptor

Kd (nM)

Bmax (pmol/mg protein)

Flag-MCHR1 i1_7sub i2_6sub i3_6sub

1.29 ± 0.42 3.25 ± 1.92 2.02 ± 1.03 2.59 ± 0.51

2.79 ± 0.69 2.62 ± 0.65 2.62 ± 0.91 2.28 ± 0.52

The membrane fractions of transiently transfected HEK293T cells were incubated with increasing concentrations of [125I] (Phe13, Tyr19) MCH from 0.02–4 nM in the absence or presence of nonlabeled MCH in an assay buffer at room temperature for 2 h. The binding reactions were terminated by rapid filtration through GF/C glass microfiber filters, and the radioactivities retained in the filters were determined using a γ-counter. Nonspecific binding was quantified in the presence of 1 μM nonlabeled MCH, and specific binding was defined as the difference between total binding and nonspecific binding. The data represent means ± SEM of three or four independent experiments performed in duplicate.

Fig. 3. Dose–response relationships of MCH-stimulated Ca2+ mobilization via FlagMCHR1 and mutant receptors. HEK293T cells transiently transfected with Flag-MCHR1 (circles) and i1_7sub (A), i2_6sub (B), or i3_6sub (C) mutants (triangles) were loaded with a non-wash calcium dye for 1 h at 37 °C. The cells were stimulated with the indicated concentrations of MCH, and the subsequent changes in the level of [Ca2+]i were measured using a Flexstation Microplate Reader. Filled symbols and continuous lines show Gi/osensitive pertussis toxin (PTX)-untreated cells, while open symbols and dashed lines show cells pretreated with 200 ng/ml PTX for 18 h. All experiments were independently performed at least three times in duplicate, and representative results are presented. The EC50 values and maximum responses are displayed in Table 3.

Table 3 Ca2+ mobilization stimulated by MCH via Flag-MCHR1 and mutant receptors. Receptor Flag-MCHR1 i1_7sub i2_6sub i3_6sub

−PTX +PTX −PTX +PTX −PTX +PTX −PTX +PTX

EC50 value of MCH (nM)

+PTX/−PTX ratio

Max response (%)

3.77 ± 0.85 12.12 ± 3.33 2.01 ± 0.50 5.38 ± 0.08 6.24 ± 2.74 8.30 ± 3.79 8.23 ± 0.28 13.68 ± 1.07

3.2

100 90.43 ± 3.41 100 94.82 ± 2.99 100 90.81 ± 2.73 100 83.31 ± 3.59

2.9 1.3a 1.7 b

HEK293T cells transfected with Flag-MCHR1 or mutants were stimulated with MCH, and the subsequent changes in the level of [Ca2+]i were measured using a Flexstation Microplate Reader. The cells were pretreated with or without 200 ng/ml Gi/o-sensitive pertussis toxin (PTX) for 18 h. The EC50 values for MCH were obtained from sigmoidal fits using a non-linear curve-fitting program. The data represent means ± SEM of three or four independent experiments performed in duplicate. a P b 0.05, significant difference from Flag-MCHR1 by Student's t-test. b P b 0.01, significant difference from Flag-MCHR1 by Student's t-test.

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

i3_6sub mutants both result in reductions in PTX sensitivity, indicating that the Gi/o activation with these two mutants was lower than that with Flag-MCHR1. As quantified by FACScan flow cytometry (Table 1), the cell surface expression level of i3_6sub was significantly reduced compared with that of Flag-MCHR1. Therefore, we examined whether the decreased level of receptor expression resulted in a reduction in Gi/o activity. We found that the Y228F/V229F mutant in TM5 showed a similar feature to i3_6sub in terms of the level of cell surface expression. However, the + PTX/− PTX ratio for Y228F/V229F was nearly equivalent to that for Flag-MCHR1 (Table 4). Thus, it is likely that the cell surface expression level did not affect the efficacy of Gi/omediated signaling. To evaluate the capacity of Gq-mediated activity, we analyzed the EC50 values in the Gi/o-uncoupled capacity in PTX-treated cells (= Gq-dependent responses) (Table 3). For all three mutants, there were no significant changes in the EC50 values for PTX-treated Gqmediated Ca2 + mobilization compared with Flag-MCHR1 (Table 3). These findings mean that the three mutants had no changes in sensitivity toward Gq activation. Taken together, in comparison with FlagMCHR1, both the i2_6sub and i3_6sub mutants showed reduced sensitivity for Gi/o, but not for Gq. This G-protein preference was also observed when i2_6sub or i3_6sub was transfected into COS-7 cells, CHO-K1 cells, and HeLa_S3 cells (data not shown). These results suggested that the region responsible for Gi/o activation with MCHR1 was not cell type-dependent. 3.3. Necessity of simultaneous substitutions in MCHR1 for decreasing the Gi/o-selective activation To identify the key residues selectively required for Gi/o activation, we continued our mutagenesis analyses based on i2_6sub and i3_6sub. Gi/o activation in the designed mutants was assessed by comparison of the ratios calculated by the + PTX/− PTX EC50 values with that of Flag-MCHR1 in Ca2+ mobilization assays. In the i2 loop, neither two double mutants (S150R/S151F and K156T/S158C) nor two quadruple mutants (S150R/S151F/T152N/K153H and S150R/K153H/K156T/ Table 4 Evaluation of Gi/o activity by comparison of +PTX/− PTX EC50 value ratios in Ca2+ mobilization. Receptor

+PTX/−PTX ratio

Flag-MCHR1

3.2

i1 loop i1_7sub

2.9

i2 loop S150R/S151F/T152N/K153H/K156T/S158C (= i2_6sub) S150R/S151F K156T/S158C S150R/S151F/T152N/K153H S150R/K153H/K156T/S158C

1.3a 3.7 2.7 5.1 2.2

i3 loop Y228F/V229F/R234N/A242L/S243P/T257M (= i3_6sub) R234N (i3 loop) T257M (i3 loop) Y228F/V229F (TM5) A242L/S243P (i3 loop) R234N/A242L/S243P/T257M (i3 loop) Y228F/V229F/R234N/S243P/T257M (TM5 + i3 loop) (=i3_5sub)

1.7 b 3.0 3.3 2.6 2.9 3.2 1.2a

HEK293T cells transfected with Flag-MCHR1 or mutants were stimulated with MCH, and the subsequent changes in the level of [Ca2+]i were measured using a Flexstation Microplate Reader. The cells were pretreated with or without 200 ng/ml Gi/o-sensitive pertussis toxin (PTX) for 18 h. The EC50 values for MCH were obtained from sigmoidal fits using a non-linear curve-fitting program. The data represent the +PTX/−PTX EC50 value ratios for Ca2+ mobilization. All experiments were independently performed at least three times in duplicate. a P b 0.05, significant difference from Flag-MCHR1 by Student's t-test. b P b 0.01, significant difference from Flag-MCHR1 by Student's t-test.

823

S158C) had altered + PTX/− PTX ratios compared with that of Flag-MCHR1 (Table 4). In the case of the i3_6sub mutant, the Y228F/ V229F mutations located in the distal region of TM5 did not reduce the +PTX/−PTX EC50 value ratio compared with that of Flag-MCHR1. Regarding the i3 loop, not only single and double mutants (R234N, T257M, and A242L/S243P) but also a quadruple mutant (R234N/ A242L/S243P/T257M) elicited similar + PTX/− PTX ratios to that of Flag-MCHR1 (Table 4). Next, we designed a mutant named i3_5sub, in which A242 in the i3 loop was left intact but two amino acids in TM5 were replaced. This mutant showed significantly decreased ratio similar to that of i3_6sub. Our finding implies that the A242L substitution in i3_6sub was not required for inhibition of Gi/o activation. We did not evaluate the phenotypes of all kinds of combined replacements in i2_6sub and i3_6sub by Ca2 + mobilization assays. However, at least, the present data can predict a necessity for multiple amino acids and the importance of two amino acids in TM5 for optimal Gi/o activation in MCHR1.

3.4. Evaluation of Gi/o-dependent activity in MCHR1 mutants by GTPγS binding assays and cAMP assays Guanine nucleotide exchange (GDP–GTP exchange) on the G-protein α-subunit is the key step in GPCR activation, and the processes can be monitored by measuring the binding of a non-hydrolyzable GTPγS analog, [35S]GTPγS. Because the Gi family G-proteins have substantially higher basal rates of GDP–GTP exchange than other G-protein families, GTPγS-binding assays are mainly a measure of GPCR-mediated activation of Gi/o proteins [37]. Therefore, to confirm the Gi/o activity, Gi/o-dependent [35S]GTPγS binding assays were performed using the membrane fractions of HEK293T cells transiently transfected with Flag-MCHR1 or mutants (Fig. 4). In cells expressing Flag-MCHR1, MCH stimulated GTPγS binding in a dose-dependent manner with an EC50 value of 0.58 ± 0.11 nM, and the maximal amount of binding with 1 μM MCH was 186.5 ± 20.6% of the basal level (means ± SEM from three independent experiments). The EC50 value for i1_7sub was 0.45 ± 0.12 nM, and the maximum response was 175.1 ± 6.0% of the basal level (Fig. 4A). Overall, there was no difference in the amounts of GTPγS binding between cells expressing Flag-MCHR1 and i1_7sub. On the other hand, both i2_6sub and i3_6sub stimulated GTPγS binding with significantly higher EC50 values of 1.30 ± 0.12 nM and 5.65 ± 2.95 nM, respectively, compared with FlagMCHR1. The maximum amounts of binding with 1 μM MCH also showed significant differences (i2_6sub: 160.2 ± 5.7%; i3_6sub: 160.2 ± 28.9%) (Fig. 4B, C). The EC50 value and maximum response for i3_5sub were almost equivalent to those for i3_6sub (i3_5sub: EC50 = 4.65 ± 1.79 nM; maximum response = 146.9 ± 23.8% of the basal level). These findings supported the notion that i3_5sub as well as i3_6sub elicited reductions in Gi/o-mediated activity. In addition, there were no significant differences in basal GTPγS binding among Flag-MCHR1, i1_7sub, i2_6sub, i3_6sub, and i3_5sub, suggesting that all four mutations did not change the constitutive receptor activity. Next, we characterized the phenotypes of i2_6sub and i3_6sub by measuring the cAMP levels after addition of MCH to stably transfected HEK293T cells. MCH potently blocked forskolin-stimulated cAMP production in Flag-MCHR1-expressing cells with an EC50 value of 0.04 ± 0.01 nM with 43.8 ± 1.3% of the maximal inhibition of cAMP accumulation (Fig. 5A). Although the cells expressing i2_6sub showed a similar extent of maximal inhibition of cAMP accumulation (46.9 ± 5.2%), the dose-response curve gave a 40-fold higher EC50 value (1.62 ± 0.39 nM, P b 0.01 versus Flag-MCHR1) (Fig. 5B). In stable i3_6sub-expressing cells, we were unable to detect any inhibition of forskolinstimulated cAMP accumulation (Fig. 5C). The basal levels and increased cAMP levels induced by forskolin in i2_6sub- or i3_6sub-transfected cells were similar to those induced in Flag-MCHR1-transfected cells. Overall, two different Gi/o-mediated assays provided evidence that

824

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

A

B

C

Fig. 4. MCH-induced [35S]GTPγS binding to Flag-MCHR1 and mutants. Transiently transfected HEK293T cells were harvested and the membrane fractions were isolated. Aliquots of the membrane fractions (10 μg) were subsequently incubated with 0.2 nM [35S]GTPγS and 0.01–1000 nM MCH in GTPγS binding buffer for 30 min at 30 °C. The bound [35S]GTPγS was separated from free [35S]GTPγS by rapid filtration through GF/C filters and washed three times. The radioactivities of the filters were counted in a scintillation cocktail using a liquid scintillation counter. The amounts of radioactivity bound to the membrane fractions are shown for Flag-MCHR1 (filled circles with continuous lines) and i1_7sub (A), i2_6sub (B), or i3_6sub (C) mutants (asterisks with dashed lines). All experiments were independently performed at least three times in triplicate, and representative results are shown.

both the i2_6sub and i3_6sub mutants caused impaired responses toward Gi/o activity.

Flag-MCHR1

i2_6sub

i3_6sub

Fig. 5. Forskolin-stimulated cAMP inhibition in Flag-MCHR1 and mutants. Inhibition of forskolin-induced cAMP accumulation is shown in HEK293T cells stably transfected with Flag-MCHR1 (A), i2_6sub mutant (B), or i3_6sub mutant (C). The cells were preincubated with assay buffer containing 0.3 mM 3-isobutyl-1-methyl-xanthine for 10 min, and then stimulated with 1 μM forskolin plus the indicated concentrations of MCH for 15 min. The reactions were terminated with 0.3 N HCl, and the levels of extracted intracellular cAMP were measured using a radioimmunoassay kit. All experiments were independently performed at least three times in duplicate, and representative results are shown.

3.5. G-protein selectivity in MCHR1 mutants evaluated by DMR assays Other than canonical second-messenger assays, we assessed Gi/o activity by DMR responses, in which receptor activity is measured as an optical trace that represents the generic response of living cells including several GPCR coupling classes. DMR assays using CHO cells transfected with representative GPCRs suggested that the ligandinduced initial robust rising response resulted from the Gi/o signature, and the ligand-induced late sustained response resulted from the Gq signature [38,39].

To examine the protein expression levels in CHO-K1 cells transiently transfected with receptors, western blotting analysis was carried out (Fig. 6A). A band of 60 kDa corresponding to Flag-MCHR1 and mutants expressing HEK293T cells was not detected but other three major bands of 35, 44 and 45 kDa were detected in CHO-K1 cells. In the cells transfected with i1_7sub or i3_6sub, the levels of mutant expression were nearly equivalent to those of Flag-MCHR1, as quantified by imaging analysis. On the other hands, the intensity of the 44/45-kDa receptor was decreased by approximately 40% in the i2_6sub mutant, compared

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

825

A kDa 64 50 37

B Gi/o

Gq

Response (pm)

Flag-MCHR1 i1_7sub

i2_6sub i3_6sub Flag-MCHR1(+PTX) Time (sec)

Flag-MCHR1 (no MCH)

Fig. 6. Dynamic mass redistribution (DMR) responses of MCH. (A) Western blotting analysis of the protein expressions of Flag-MCHR1 and individual mutant receptors. Transiently transfected CHO-K1 cells were lysed with SDS-sample buffer, and then homogenized at 4 °C by sonication. The proteins were separated by 12.5% SDS-PAGE, transferred to a PVDF membrane, and immunoblotted with an anti-DYKDDDDK antibody. Three major immunoreactive bands of 35, 44, and 45 kDa are present in Flag-MCHR1 and individual mutant receptors, while the intensity of the 44/45-kDa receptor (arrow) appears to be reduced in the i2_6sub mutant. (B) CHO-K1 cells transiently transfected with Flag-MCHR1 or mutants were pretreated with or without 200 ng/ml pertussis toxin (PTX) for 16–20 h. The cells were washed four times with HBSS/20 mM HEPES, and incubated for 2 h at room temperature. After a steady baseline was established, the ligand solutions were added to individual wells. The real-time DMR signals of the cells were recorded in response to stimulation with or without 1 μM MCH. All experiments were independently performed at least three times in triplicate, and representative results are shown.

with that for Flag-MCHR1. As described above, the G-protein preference was observed by measurements of intracellular Ca2 + mobilization when i2_6sub or i3_6sub was transfected into CHO-K1 cells. Similarly with respect to HEK293T, it is likely that the receptor expression level in CHO-K1 cells did not affect the efficacy of Gi/o-mediated signaling. Thus, we chose to use the CHO-K1 cells for assessing Gi/o activity by DMR responses. As expected, addition of MCH to CHO-K1 cells transiently transfected with Flag-MCHR1 induced a combined optical trace triggered from Gq and Gi/o proteins. Thus, the DMR response was composed of two phases: one as an initial robust rising response caused by Gi/o-

dependent activity and the other as a late sustained response caused by Gq-dependent activity (Fig. 6B). Indeed, the robust rising response was exclusively decreased in Gi/o-sensitive PTX-treated cells transfected with Flag-MCHR1. While the DMR trace obtained from i1_7sub was similar to that obtained from Flag-MCHR1, i2_6sub- and i3_6sub-expressing cells exhibited reductions in the Gi/o-mediated robust rising response. We further quantified a Gi/o-dependent maximal DMR response by measuring the DMR peak amplitude during 300 s post MCH stimulation. In cells expressing Flag-MCHR1, the maximal amplitude in PTX-untreated cells was 73.82 ± 3.54 picometer (pm), while that in PTX-treated cells was 31.92 ± 4.74 pm. The maximal

826

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

amplitude for i1_7sub, 79.99 ± 7.46 pm, was almost equivalent to that for Flag-MCHR1. On the other hand, both i2_6sub and i3_6sub showed significantly lower Gi/o-dependent activity with maximal amplitude of 59.50 ± 3.16 pm and 46.09 ± 1.41 pm, respectively, in comparison to Flag-MCHR1. These results further confirm that i2_6sub and i3_6sub evoked impaired Gi/o activity compared with Flag-MCHR1. 4. Discussion Despite the large amount of information available on the structure– function relationships of GPCRs, very little is known about the principles and determinants of G-protein coupling. Identification of structural determinants for G-protein selectivity has been performed by alanine scanning mutagenesis or model-driven mutagenesis. In this study, we employed a different type of approach based on the findings in our previous studies. Specifically, using a heterologous cell expression system, we found that rat MCHR1 couples to both Gi/o and Gq, whereas goldfish MCHR1 exclusively couples to Gq [5,35,36]. Therefore, we took advantage of this interspecies difference to predict amino acid residues in rat MCHR1 that may be selectively responsible for Gi/o coupling. By comparing sequence alignments between rat and goldfish MCHR1s, three multisubstituted mutants of rat MCHR1 containing the fish-type residue substitutions were designed. Through analyses with 10 other substitution mutants using Ca2 + mobilization assays (Table 4), two MCHR1 mutants, i2_6sub and i3_6sub, were found to exhibit reduced Gi/o activities without altering Gq-mediated signaling. The i2_6sub mutant had six simultaneously substituted residues in the i2 loop, while i3_6sub had four simultaneously substituted residues in the i3 loop and two residues in TM5. Finally, measurements of other signaling events with the feature of total cellular responses confirmed the phenotypes of impaired Gi/o activation in i2_6sub and i3_6sub. The remarkable differences in the lengths and sequences of the i2 and i3 loops in GPCRs have been inferred to regulate receptor selectivity to different G-proteins [29–31]. However, only a few lines of evidence have indicated that the i2 loop in GPCRs contributes to the receptor selectivity with Gq-protein interactions. For example, in proteaseactivated receptor 1 belonging to the class A GPCRs, an R205A mutation located in the proximal part of the i2 loop resulted in a selective decrease in Gq/11 activity without any effects on Gi/o- and G12/13mediated signaling [34]. In a class B GPCR, parathyroid hormone (PTH)/PTH-related protein receptor, a K-to-E mutation located in the middle region of the i2 loop resulted in normal Gs-mediated cAMP responses, but impaired Gq-mediated phospholipase C activation [40]. Therefore, our i2_6sub mutant (S150R/S151F/T152N/K153H/K156T/ S158C) is the first reported mutant to provide evidence for the importance of the i2 loop for a selective decrease in Gi/o activity. Interestingly, our previous study demonstrated that Arg155 adjacent to Lys156 is a recognition determinant with relevance to both Gi/o- and Gqmediated pathways in rat MCHR1 without changing the level of cell surface expression [23]. These results suggest that each amino acid located in the distal part of the i2 loop may have different roles in determining receptor activity and G-protein sensitivity in rat MCHR1. Regarding the i3 loop, mutations of the two membrane distal basic amino acids (Lys and Arg) to Ala in α2A-adrenergic receptor impaired cAMP inhibition via Gi/o, while cAMP activation via Gs was unaffected [41]. Another example is leukotriene B4 receptor 1. By tandem alanine replacement of three to four amino acids in the i1, i2, and i3 loops, DIGR located in the proximal i3 loop was identified to be selectively involved in Gi/omediated, but not G16-mediated, signaling [32]. In our i3_6sub mutant (Y228F/V229F/R234N/A242L/S243P/T257M), the distal basic amino acid was not substituted and the tandem DIGR sequence does not exist in the i3 loop of rat MCHR1. Furthermore, substitutions in TM5 in addition to the i3 loop were necessary to elicit optimal Gi/o-mediated signaling in MCHR1 (Table 4). Thus, i2_6sub and i3_6sub are novel types of GPCR mutants that are selectively involved in Gi/o-mediated signaling.

It has been postulated that a spatial conformational feature of GPCRs is responsible for selectivity for certain G-protein subtypes, because different activating agonists can cause different G-protein-subtype preferences for a single receptor [42]. As noted previously, a threedimensional homology model of rat MCHR1 predicted that the i2 and i3 loops are exposed and face away from the interior of the structure, and thus cannot be involved in other cytoplasmic intracellular loop interactions [23,43]. Therefore, it is assumed that the i2 and i3 loops are easily accessible to G-proteins. Hence, it can be speculated that the i2_6sub and i3_6sub mutants of MCHR1 significantly destabilize the spatial interactions with Gi/o, but not Gq. Alternatively, i2_6sub and i3_6sub could lose particular interaction patterns with Gi/o protein that permit nonspecific interactions with other molecules. As evidenced by our analyses from the mutant cassette (Table 4), we could not address a specific amino acid or consecutive amino acids that were responsible for Gi/o selectivity. Eventually, the complexity derived from multisubstituted residues prevents second and third structural predictions by building molecular models of the mutants with Gi/o heterotrimers. Further studies are needed to clarify the key determinants by various combinations in which individual mutant amino acids in i2_6sub and i3_6sub are reverted to the wild-type residues in rat MCHR1. This would allow us to solve how the substitutions in i2_6sub and i3_6sub affect the potential interface and the recognition features with Gi/o protein. Based on our findings regarding several determinants within MCHR1 toward Gα-protein coupling, we need to address how MCHR1 couples to Gα-proteins in a natural cellular context. Mammalian MCHR1 transfected into heterologous expressing cells is coupled to both Gi/o and Gq, albeit with a higher affinity for Gi/o than for Gq [5, 19]. Regarding studies on cell lines with endogenous MCHR1 expression, MCHR1 was exclusively coupled to Gi/o in the human melanoma cell SK-MEL-37 and human neuroblastoma cell Kelly [44,45]. On the other hand, MCH predominantly caused Gq-mediated signaling in the human neuroblastoma cell IMR32 [46], while MCHR1 enabled the synthesis of cAMP probably via Gs in peripheral blood mononuclear cells [47]. In a physiological context, strong inhibition of synaptic activity in lateral hypothalamus neurons was caused by MCH-mediated signaling bias toward Gi/o on MCHR1 [48]. Thus, the cellular environment in which the receptor is expressed leads to preferential enhancement of selective Gα-protein-mediated signaling by MCHR1. The signaling via i2_6sub and i3_6sub is in favor of the Gq side. Therefore, we can propose to establish either i2_6sub or i3_6sub knockin mice, in which the total signaling balance is manipulated in favor of the Gq side through MCHR1. Because MCHR1 is highly expressed in several regions such as the hippocampus, amygdala, and nucleus accumbens shell [8], it is likely that a distinct brain region may exhibit a selective Gq-Gi/o balance via MCHR1. Using such a possibility, it will be interesting to modify the Gq-Gi/o balance in individual brain regions achieved by virusmediated overexpression of either i2_6sub or i3_6sub. Examination of the behavioral outcomes in either i2_6sub or i3_6sub engineered rodents could have implications for understanding the potential physiological role of the balance between Gi and Gq signaling via MCHR1. 5. Conclusions Because the balance between Gi and Gq signaling has been suggested to regulate antipsychotic or propsychotic activity in vivo [49], it is important to analyze the selectivity mechanisms between GPCR and different subtypes of G-proteins. In this study, we have identified two mutants within the i2 loop or i3 loop/TM5 of rat MCHR1 that exhibited significant decreases in Gi/o-mediated, but not Gq-mediated signaling. Therefore, our mutants may be useful for assessing a series of biased ligand that selectively induce cellular signaling via Gq protein. Overall, our data highlight a previously unknown molecular mechanism that rat MCHR1 possesses two different intermolecular regions that act a switch for activation of Gi/o protein. Further research is required to

A. Hamamoto et al. / Cellular Signalling 27 (2015) 818–827

identify the key amino acid residues in i2_6sub and i3_6sub, and how they activate the MCHR1 conformation in complex with Gi/o heterotrimers. Our present data on specific interaction between MCHR1 and Gi/o proteins will hopefully allow additional understanding of dual G-protein coupling in other class A GPCRs. Authors' contributions Y.S. and A.H. conceived and designed the experiments; A.H. conducted the experiments; Y.K. contributed to the experiments involving cAMP assays and DMR assays; Y.S. and A.H. analyzed the data and wrote the manuscript. Conflict of interests The authors declare that they have no competing interests. Acknowledgments This study was supported by research grants from the Ministry of Education, Culture, Sports, and Technology of Japan (KAKENHI 20500337 and 23500449 to Y.S.) and a Grant-in-Aid for JSPS Fellows (13J01969 to A.H.) from the Japan Society for the Promotion of Science. References [1] J.C. Bittencourt, F. Presse, C. Arias, C. Peto, J. Vaughan, J.L. Nahon, W. Vale, P.E. Sawchenko, J. Comp. Neurol. 319 (1992) 218–245. [2] D. Qu, D.S. Ludwig, S. Gammeltoft, M. Piper, M.A. Pelleymounter, M.J. Cullen, W.F. Mathes, R. Przypek, R. Kanarek, E. Maratos-Flier, Nature 380 (1996) 243–247. [3] M. Shimada, N.A. Tritos, B.B. Lowell, J.S. Flier, E. Maratos-Flier, Nature 396 (1998) 670–674. [4] J. Chambers, R.S. Ames, D. Bergsma, A. Muir, L.R. Fitzgerald, G. Hervieu, G.M. Dytko, J.J. Foley, J. Martin, W.S. Liu, J. Park, J.C. Ellis, S. Ganguly, S. Konchar, J. Cluderay, R. Leslie, S. Wilson, H.M. Sarau, Nature 400 (1999) 261–265. [5] Y. Saito, H.P. Nothacker, Z. Wang, S. Lin, F.M. Leslie, O. Civelli, Nature 400 (1999) 265–269. [6] S. An, G. Cutler, J.J. Zhao, S.G. Huang, H. Tian, W. Li, L. Liang, M. Rich, A. Bakleh, J. Du, J.L. Chen, K. Dai, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 7576–7581. [7] C.P. Tan, H. Sano, H. Iwaasa, J. Pan, A.W. Sailer, D.L. Hreniuk, S.D. Feighner, O.C. Palyha, S.S. Pong, D.J. Figueroa, C.P. Austin, M.M. Jiang, H. Yu, J. Ito, M. Ito, M. Ito, X.M. Guan, D.J. MacNeil, A. Kanatani, L.H. Van der Ploeg, A.D. Howard, Genomics 79 (2002) 785–792. [8] Y. Saito, M. Cheng, F.M. Leslie, O. Civelli, J. Comp. Neurol. 435 (2001) 26–40. [9] Y. Chen, C. Hu, C.K. Hsu, Q. Zhang, C. Bi, M. Asnicar, H.M. Hsiung, N. Fox, L.J. Slieker, D.D. Yang, M.L. Heiman, Y. Shi, Endocrinology 143 (2002) 2469–2477. [10] D.J. Marsh, D.T. Weingarth, D.E. Novi, H.Y. Chen, M.E. Trumbauer, A.S. Chen, X.M. Guan, M.M. Jiang, Y. Feng, R.E. Camacho, Z. Shen, E.Z. Frazier, H. Yu, J.M. Metzger, S.J. Kuca, L.P. Shearman, S. Gopal-Truter, D.J. MacNeil, A.M. Strack, D.E. MacIntyre, L.H. Van der Ploeg, S. Qian,, D.E. MacIntyre, L.H. Van der Ploeg, S. Qian, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 3240–3245. [11] S. Takekawa, A. Asami, Y. Ishihara, J. Terauchi, K. Kato, Y. Shimomura, M. Mori, H. Murakoshi, K. Kato, N. Suzuki, O. Nishimura, M. Fujino, Eur. J. Pharmacol. 438 (2002) 129–135. [12] L.P. Shearman, R.E. Camacho, D. Sloan Stribling, D. Zhou, M.A. Bednarek, D.L. Hreniuk, S.D. Feighner, C.P. Tan, A.D. Howard, L.H. Van der Ploeg, D.E. MacIntyre, G.J. Hickey, A.M. Strack, Eur. J. Pharmacol. 475 (2003) 37–47. [13] B. Borowsky, M.M. Durkin, K. Ogozalek, M.R. Marzabadi, J. DeLeon, B. Lagu, R. Heurich, H. Lichtblau, Z. Shaposhnik, I. Daniewska, T.P. Blackburn, T.A. Branchek, C. Gerald, P.J. Vaysse, C. Forray, Nat. Med. 8 (2002) 825–830.

827

[14] D. Georgescu, R.M. Sears, J.D. Hommel, M. Barrot, C.A. Bolaños, D.J. Marsh, M.A. Bednarek, J.A. Bibb, E. Maratos-Flier, E.J. Nestler, R.J. DiLeone, J. Neurosci. 25 (2005) 2933–2940. [15] R. Fredriksson, H.B. Schiöth, Mol. Pharmacol. 67 (2005) 1414–1425. [16] H. Kawauchi, I. Kawazoe, M. Tsubokaw, M. Kishida, B.L. Baker, Nature 305 (1983) 321–323. [17] K. Matsuda, K. Kojima, S. Shimakura, A. Takahashi, Peptides 30 (2009) 2060–2065. [18] Y. Kobayashi, A. Hamamoto, T. Hirayama, Y. Saito, Gen. Comp. Endocrinol. (2014). http://dx.doi.org/10.1016/j.ygcen.2014.03.005. [19] B.E. Hawes, E. Kil, B. Green, K. O'Neill, S. Fried, M.P. Graziano, Endocrinology 141 (2000) 4524–4532. [20] Y. Aizaki, K. Maruyama, M. Nakano-Tetsuka, Y. Saito, Peptides 30 (2009) 974–981. [21] Y. Saito, M. Tetsuka, L. Yue, Y. Kawamura, K. Maruyama, FEBS Lett. 533 (2003) 29–34. [22] D. Macdonald, N. Murgolo, R. Zhang, J.P. Durkin, X. Yao, C.D. Strader, M.P. Graziano, Mol. Pharmacol. 58 (2000) 217–225. [23] Y. Saito, M. Tetsuka, S. Saito, K. Imai, A. Yoshikawa, H. Doi, K. Maruyama, Endocrinology 146 (2005) 3452–3462. [24] J. Fan, S.J. Perry, Y. Gao, D.A. Schwarz, R.A. Maki, Mol. Endocrinol. 19 (2005) 2579–2590. [25] M. Tetsuka, Y. Saito, K. Imai, H. Doi, K. Maruyama, Endocrinology 145 (2004) 3712–3723. [26] Y. Saito, M. Tetsuka, Y. Li, H. Kurose, K. Maruyama, Peptides 25 (2004) 1597–1604. [27] Y. Saito, A. Hamamoto, Y. Kobayashi, Front. Endocrinol. (Lausanne) 4 (2013) 154. [28] A. Hamamoto, M. Horikawa, T. Saho, Y. Saito, Front. Endocrinol. (Lausanne) 26 (2012) 147. [29] E. Hermans, Pharmacol. Ther. 99 (2003) 25–44. [30] V. Katritch, V. Cherezov, R.C. Stevens, Trends Pharmacol. Sci. 33 (2012) 17–27. [31] I.S. Moreira, Biochim. Biophys. Acta 1840 (2014) 16–33. [32] K. Kunieda, T. Okuno, K. Terawaki, M. Miyano, T. Yokomizo, T. Shimizu, J. Biol. Chem. 282 (2007) 3998–4006. [33] G. Kleinau, H. Jaeschke, C.L. Worth, S. Mueller, J. Gonzalez, R. Paschke, G. Krause, PLoS ONE 5 (2010) e9745. [34] K.L. McCoy, S. Gyoneva, C.P. Vellano, A.V. Smrcka, S.F. Traynelis, J.R. Hepler, Cell. Signal. 24 (2012) 1351–1360. [35] K. Mizusawa, Y. Saito, Z. Wang, Y. Kobayashi, T. Matsuda, A. Takahashi, Peptides 30 (2009) 1990–1996. [36] A. Hamamoto, K. Mizusawa, A. Takahashi, Y. Saito, Regul. Pept. 169 (2011) 6–12. [37] G. Milligan, Trends Pharmacol. Sci. 24 (2003) 87–90. [38] R. Schröder, N. Janssen, J. Schmidt, A. Kebig, N. Merten, S. Hennen, A. Müller, S. Blättermann, M. Mohr-Andrä, S. Zahn, J. Wenzel, N.J. Smith, J. Gomeza, C. Drewke, G. Milligan, K. Mohr, E. Kostenis, Nat. Biotechnol. 28 (2010) 943–949. [39] L. Sundström, P.J. Greasley, S. Engberg, M. Wallander, E. Ryberg, FEBS Lett. 587 (2013) 2399–2404. [40] A. Iida-Klein, J. Guo, M. Takemura, M.T. Drake, J.T. Potts Jr., A. Abou-Samra, F.R. Bringhurst, G.V. Segre, J. Biol. Chem. 272 (1997) 6882–6889. [41] S.M. Wade, W.K. Lim, K.L. Lan, D.A. Chung, M. Nanamori, R.R. Neubig, Mol. Pharmacol. 56 (1999) 1005–1013. [42] S.K. Wong, Neurosignals 12 (2003) 1–12. [43] A. Nagata, A. Hamamoto, M. Horikawa, K. Yoshimura, S. Takeda, Y. Saito, Gen. Comp. Endocrinol. 188 (2013) 159–165. [44] Y. Saito, Z. Wang, K. Hagino-Yamagishi, O. Civelli, S. Kawashima, K. Maruyama, Biochem. Biophys. Res. Commun. 289 (2001) 44–50. [45] S.E. Schlumberger, V. Jäggin, H. Tanner, A.N. Eberle, Biochem. Biophys. Res. Commun. 298 (2002) 54–59. [46] D. Fry, B. Dayton, S. Brodjian, C. Ogiela, H. Sidorowicz, L.J. Frost, T. McNally, R.M. Reilly, C.A. Collins, Int. J. Biochem. Cell Biol. 38 (2006) 1290–1299. [47] M. Verlaet, A. Adamantidis, B. Coumans, G. Chanas, W. Zorzi, E. Heinen, T. Grisar, B. Lakaye, FEBS Lett. 527 (2002) 205–210. [48] X.B. Gao, A.N. van den Pol, J. Physiol. 533 (2001) 237–252. [49] M. Fribourg, J.L. Moreno, T. Holloway, D. Provasi, L. Baki, R. Mahajan, G. Park, S.K. Adney, C. Hatcher, J.M. Eltit, J.D. Ruta, L. Albizu, Z. Li, A. Umali, J. Shim, A. Fabiato, A.D. Jr MacKerell, V. Brezina, S.C. Sealfon, M. Filizola, J. González-Maeso, D.E. Logothetis, Cell 147 (2011) 1011–1023.