Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jan 7.
Published in final edited form as: Structure. 2013 Nov 21;22(1):149–155. doi: 10.1016/j.str.2013.10.002

Functional assay for T4 lysozyme-engineered G Protein-Coupled Receptors with an ion channel reporter

Katarzyna Niescierowicz 1,2,3, Lydia Caro 1,2,3,4, Vadim Cherezov 5, Michel Vivaudou 1,2,3, Christophe J Moreau 1,2,3,*
PMCID: PMC3947161  NIHMSID: NIHMS532273  PMID: 24268646

Abstract

Summary:

Structural studies of G protein-coupled receptors (GPCRs) extensively use the insertion of globular soluble protein domains in order to facilitate their crystallization. However, when inserted in the third intracellular loop (i3 loop), the soluble protein domain disrupts their coupling to G proteins and impedes the GPCRs functional characterization by standard G protein-based assays. Therefore, activity tests of crystallization-optimized GPCRs are essentially limited to their ligand binding properties using radioligand binding assays. Functional characterization of additional thermostabilizing mutations requires the insertion of similar mutations in the wild-type receptor to allow G protein-activation tests. We demonstrate that Ion Channel-Coupled Receptor technology is a complementary approach for a comprehensive functional characterization of crystallization-optimized GPCRs and potentially of any engineered GPCR. Ligand-induced conformational changes of the GPCRs are translated into electrical signal and detected by simple current recordings, even though binding of G proteins is sterically blocked by the added soluble protein domain.

Introduction

G protein-coupled receptors (GPCRs) are membrane proteins involved in cellular communications and environment sensing in eukaryotic organisms. The specific binding of circulating ligands is transduced inside the cell and amplified by intracellular pathways leading to an adapted cellular response. Due to their central role in human physiology as well as their abundance (>800 human genes), GPCRs are major pharmaceutical targets. With the dual objective of developing new pharmaceutical compounds and understanding the complex molecular mechanisms of signal transduction, high-resolution crystallographic structures of GPCRs provide invaluable data (Katritch et al., 2013). However, low expression level and intrinsic instability of GPCRs have proven to be major impediments to crystallographic studies. Thus, the rhodopsin receptor, which is naturally abundant in the photoreceptor cell outer segments and stabilized in an inactive dark state by a covalently bound ligand and an ionic lock, was the first GPCR structure determined, and remained the only one for a further 7 years until the structure of the β2-adrenergic receptor was solved (Venkatakrishnan et al., 2013). To solve GPCR structures other than rhodospin, three successful methods have been used, alone or in combination, to overcome receptor instability and increase the solvent-exposed surface area available for crystal contacts: (i) co-crystallization with stabilizing antibodies (Rasmussen et al., 2007), including nanobodies (Rasmussen et al., 2011); (ii) insertion of T4 phage lysozyme (T4L) (Cherezov et al., 2007; Rosenbaum et al., 2007), thermostabilized apocytochrome b562RIL (BRIL) (Chun et al., 2012; Liu et al., 2012), or rubredoxin (Tan et al., 2013) domains; (iii) introduction of thermostabilizing mutations (Warne et al., 2008). The T4L insertion in the i3 loop has been so far the most successful strategy with 14 distinct receptors crystallized in different states (Katritch et al., 2013). Unfortunately, the insertion of a T4L or other soluble protein fusion domain in the i3 loop disrupts interactions with G proteins and hinders functional characterization of the modified receptors with standard G protein-dependent signalling assays (Liu et al., 2012; Wu et al., 2010). Testing the function of engineered receptors is therefore limited to radioligand binding and competition assays. However, these assays evaluate only the proper folding of the ligand binding sites by comparing affinities with the wild-type (wt) receptor. In the case of the β2-adrenergic receptor, functional characterization was completed by fluorescence spectroscopy of the bimane-labeled Cys2656.27 β2(T4L) (Rosenbaum et al., 2007), demonstrating ligand-induced conformational changes of the cytoplasmic end of helix VI. These results were similar to those observed with the wt β2-adrenergic receptor. This method of fluorescence spectroscopy has not been used with other crystallized GPCRs having a T4L or other fusion domain. In most cases, the constructs providing the highest crystallographic resolution required adjustments of the fusion partner insertion sites (Kruse et al., 2012), which could affect the receptor dynamics and lead to partially active or inactive states. Occasionally, thermostabilizing mutations are incorporated in the T4L-modified receptors in order to increase the resolution of crystallographic structures. Such mutations can affect the activity of GPCRs and their impact must be tested, usually by inserting them in the wt receptor to perform standard G protein-based assays (White et al., 2012).

Building on the concept of Ion Channel-Coupled Receptors (ICCRs) (Caro et al., 2011; Caro et al., 2012; Moreau et al., 2008), we developed a functional assay for GPCRs by genetically fusing a potassium channel to the receptor C-terminus. The channel acts as a direct reporter of the agonists- and antagonists-induced conformational changes of the GPCRs. Upon binding of agonists, the receptor undergoes conformational changes propagating from the external ligand binding site to the intracellular side leading to the activation of G proteins. In ICCRs, agonist binding induces either an activation (M2 and β2 ICCRs) or an inhibition (D2 and Opsin ICCRs) of the fused channel. Antagonists are detected by their ability to block the agonist effect. Using G protein-blocking toxins, current recordings in cell-free environment and co-expression of unfused receptors and channels, we demonstrated that the fused channel serves as a reporter of the activated state of the GPCR (Moreau et al., 2008). Conformational changes induced by activation of the GPCR are transmitted to the channel gate(s) resulting in stabilization of the channel in open-state (activation) or in closed-state (inhibition). Concentration-effect curves with agonists revealed a correlation between the amplitude of the current and the concentration of agonists, indicating that the affinity and the efficacy of a ligand finely tuned the kinetics of channel gating. The generated electrical signal is measured by electrophysiological methods providing the following advantages: (i) low-cost and non-radioactive reagents; (ii) real-time recordings on single cells; (iii) detection of agonists and antagonists in concentration-dependent manner; (iv) functional characterization of additional thermostabilizing mutations in the T4L constructs. Most notably, the ICCR technology is independent of G protein activation and simplifies the functional characterization of engineered G protein-(un)coupled receptors and additional mutations. In this study we demonstrate the potential of ICCR technology to functionally characterize GPCRs optimized for crystallization by T4L insertion in the i3 loop.

Results

Proof-of-concept: M2(T4L) functional characterization

The ability of ICCRs to functionally characterize GPCRs(T4L) optimized for crystallization was first tested with the human M2 muscarinic receptor (Fig. 1A). Starting from the human M2 receptor fused to Kir6.2, the T4L domain was genetically inserted to produce a receptor identical to the crystallized M2(T4L) (Haga et al., 2012) with the exception of 2 differences: (i) the glycosylation sites were not removed; and (ii) the last 9 residues were removed creating an agonist-inhibited ICCR.

Figure 1. Electrophysiological characterization of the M2(T4L) muscarinic receptor.

Figure 1

Open in a new tab

(A) Schematic representation of ICCR(T4L) showing, in the plane of the membrane, one of the 4 subunits of the ICCR and the Kir6.2 part of the facing subunit. Ligand-induced conformational changes of T4L-modified GPCRs are transduced into electrical signals by the fused Kir6.2 channel. KΔ represents the Kir6.2ΔC36. (B) Whole-cell basal current amplitudes reflect the surface expression levels of M2 ICCRs. Bars represent the mean +/− SEM of whole-cell currents. The dashed line represents the average endogenous current recorded in non-injected oocytes. Numbers of recordings are indicated above bars. (C) Representative TEVC recordings at −50 mV showing M2 ICCR responses to 5 μM ACh. Grey and blue arrows indicate basal current and channel inhibition during ACh application, respectively. Barium (Ba2+, 3mM) is a generic blocker of potassium channels used to determine the baseline (dashed line). (D) The antagonist atropine (1 μM, red line and arrow) inhibits the effect of ACh. (E) Percent inhibition induced by 5 μM ACh (blue bar) and 5 μM Ach + 1 μM atropine (red bar). Percentages are calculated in reference to the current amplitude before ligand applications. Bars represent mean +/− SEM. n=6.

ICCRs were heterologously expressed in Xenopus oocytes by microinjection of mRNA. Whole-cell currents were recorded by manual or automated two-electrode voltage-clamp (TEVC) in symmetrical high K+ concentrations. The surface expression of ICCRs was roughly estimated by the amplitude of the basal current in the absence of ligands. Results presented in figure 1B reveal that the insertion of T4L unexpectedly abolished surface expression of the muscarinic ICCR. Physiologically, surface expression of Kir6.2 requires a physical interaction with the sulfonylurea receptor, its natural partner, in order to mask an endoplasmic reticulum (ER) retention signal, the Arg-Lys-Arg sequence, present in the channel cytoplasmic C-terminus[ (Zerangue et al., 1999). Removal of this signal by truncation of the last 36 residues of Kir6.2, denoted KΔ, allows surface expression of the channel alone (Tucker et al., 1997). We found that such removal in the ICCR(T4L) was beneficial and the construct M2(T4L)-KΔ displayed high surface expression (Fig. 1B). This is in contrast with our previous observation that the presence of the ER retention signal did not preclude high surface expression of M2 ICCRs (Moreau et al., 2008), suggesting that the insertion of the T4L domain in M2 receptor affects the masking of the channel ER retention signal.

Ligand-induced conformational changes are detected by the change of current generated by the fused Kir6.2 channel. In ICCRs, the open probability of Kir6.2 is similar to that of the unfused channel (KΔ) and the channel is partly open at rest in our experimental conditions. This basal activity offers the advantage of sensing either agonist-evoked channel inhibition (shortened M2, rhodopsin and D2L ICCRs) (Caro et al., 2012; Moreau et al., 2008) or activation (full-length M2, β2 ICCRs) (Caro et al., 2011; Moreau et al., 2008). In the case of the M2(T4L) ICCR, the physiological agonist acetylcholine (ACh) causes an inhibition of Kir6.2 comparable to that observed with the wt M2 ICCR (Fig. 1C). This result demonstrates the ability of this system to detect agonist-induced conformational changes of M2(T4L) receptor by simple current recording. Antagonists are also detected by M2 and M2(T4L) ICCRs, as shown in figures 1D&E with atropine (1 μM) being concomitantly applied with ACh (5 μM). Functional characterization of M2(T4L) receptor was completed with concentration-response curves (Fig. 2A) for carbachol (CCh), a synthetic agonist, showing that T4L insertion does not change the affinity reported by the ICCR assay. This result is in agreement with competitive radioligand binding assays performed on M2 and M2(T4L) receptors (Haga et al., 2012). We note that the insertion of T4L reduced the efficacy of the agonist, as shown by the lower amplitude of the ICCR signal.

Figure 2. Role of the i3 loop in the efficacy of agonist on M2 ICCRs.

Figure 2

Open in a new tab

(A) Concentration-dependent inhibition of M2 ICCRs by carbachol (CCh, an ACh analogue). Increasing concentrations of CCh are applied on the same cell expressing the indicated constructs. M2 + KΔ corresponds to the co-expression of the unfused M2 receptor and Kir6.2ΔC36 channel. Δi3 signifies deletion of the i3 loop. Values are mean +/− SEM. Negative values indicate an inhibition of the current generated by Kir6.2. n≥7. (B) Whole-cell basal currents including M2(Δi3)-K ICCR. Bars represent the mean +/− SEM. Numbers of recordings are indicated above bars. (C) Co-expression of the G protein-activated Kir3.4* channel with the i3 loop-lacking ICCR, M2(Δi3)-K. Due to a higher surface expression level of the Kir3.4* channel, the current amplitude generated by this channel is much larger than the current generated by Kir6.2. TEVC recordings show that binding of ACh (5 μM) to the fused M2(Δi3) receptor activates the Kir3.4* channels demonstrating that this modified receptor is able to activate Gi/o proteins. (D) Mean +/− SEM of the percentage of activation of Kir3.4* induced by 5 μM Ach on cells expressing M2 and M2(Δi3) ICCRs.

Engineering the GPCRs with the T4L domain generates 2 modifications: (i) deletion of the i3 loop; (ii) insertion of the T4L domain. In order to discriminate between the effect of i3 loop deletion and that of the T4L addition on the lower efficacy of the agonist on the M2(T4L) ICCR, we created a M2 ICCR without the i3 loop (M2(Δi3)-K) and performed a CCh-concentration effect curve. Figure 2A shows a similar curve for the T4L and the Δi3 constructs indicating that the lower observed efficacy is due to the deletion of the i3 loop and not to the addition of the T4L domain. In contrast, surface expression of M2(Δi3)-K, despite the presence of the ER retention signal (Fig. 2B), indicates that the lack of expression of M2(T4L)-K is due to the insertion of the T4L domain and not to the deletion of the i3 loop. The insertion of the T4L domain in the i3 loop disrupts the interaction with G proteins as observed for all crystallized GPCR(T4L), while the deletion of the i3 loop preserves the ability of the receptor to activate G proteins, as verified by the activation of G protein-activated Kir3.4 channels (Fig. 2C&D). This observation raises the hypothesis that surface expression of M2 ICCRs requires interaction with partners such as G proteins able to mask the Kir6.2 ER retention signal and that this interaction is disrupted by the T4L domain when inserted in the i3 loop.

Validation with another GPCR: β2(T4L) ICCR

We demonstrated that ICCR technology is able to functionally characterize M2(T4L) receptor. In order to validate this concept with another GPCR, we used the previously created adrenergic β2-K ICCR (Caro et al., 2011) and inserted the T4L domain in the i3 loop. The position of the T4L domain was the same as in the first crystallized β2(T4L) receptor (Rosenbaum et al., 2007), but the glycosylation sites were left intact. Removal of the Kir6.2 ER retention signal was required to detect the β2(T4L)-K ICCR at the cell surface, as observed for the M2 ICCR (Fig. 3B). The β2 ICCRs have the particularity to require the co-expression of the transmembrane domain 0 (TMD0) from the sulfonylurea receptor 1 (SUR1) (Caro et al., 2012), and its presence is implicit with all β2 ICCRs. TMD0 tightly binds to the Kir6.2 channel (Chan et al., 2003), presumably acting as a chaperone on ICCRs.

Figure 3. Characterization of another GPCR(T4L): the β2(T4L) adrenergic receptor.

Figure 3

Open in a new tab

(A) Schematic representations of β2(T4L) ICCR. (B) Whole-cell basal currents generated by the indicated constructs. Implicitly, all β2 ICCRs were co-expressed with the N-terminal domain of the sulfonylurea receptor 1 (TMD0) to boost their surface expression as previously described14. Values are mean +/− SEM. Numbers above bars represent the number of oocytes tested. (C)

TEVC recordings showing the activation of the fused Kir6.2 (brown arrows) by the adrenergic agonist isoproterenol 0.5 μM in cells expressing β2 and β2(T4L) ICCRs. (D) Percentage of activation of β2 constructs by 0.5 μM isoproterenol. Values are mean +/− SEM and the number of experiments is indicated over the bars. (E) Concentration-effect curves of isoproterenol on the indicated β2 ICCRs and the control with the unfused receptor and channel (β2 + KΔ). Values are mean +/− SEM. n ≥ 5.

In this configuration, the activation of β2 receptor by 0.5 μM isoproterenol, an adrenergic agonist, leads to the activation of Kir6.2 (Chan et al., 2003). Applied on the β2(T4L)-K ICCR (Fig. 3C&D), the agonist induced the same response, validating the ICCR functional assay with a second GPCR(T4L). Comparison of the agonist concentration-effect curves of β2 and β2(T4L) receptors (Fig. 3E) confirmed the results obtained with M2(T4L) ICCR: similar apparent affinities but lower efficacy of the T4L construct.

Extrapolation to a not-yet crystallized GPCR: OXTR(T4L) ICCR

To validate this method with a 3rd receptor not yet crystallized, we chose the oxytocin receptor (OXTR) as a model (Fig. 4A). This receptor has been optimized for crystallization with the following modifications: (i) i3 loop replacement by the T4L domain; (ii) C-terminal truncation of the last 42 residues; and (iii) gene optimization for expression in insect cells. As for M2 and β2 receptors, surface expression of the OXTR(T4L) ICCR required the elimination of the Kir6.2 ER retention signal (Fig. 4B), confirming the deleterious effect of the T4L domain on the surface expression of ICCRs.

Figure 4. Application of the ICCR functional assay to a not-yet-crystallized GPCR: the oxytocin receptor.

Figure 4

Open in a new tab

(A) Diagram of oxytocin ICCR OXTR(T4L)-K. (B) Assessment of the surface expression of the OXTR(T4L) ICCR by measuring the whole-cell basal current. Bars represent mean +/− SEM and numbers above bars, the number of recordings. (C) TEVC recordings showing the activation of the ICCR during application of the agonist oxytocin 1 μM. (D) Statistical results represented as mean +/− SEM of percentage of change in current induced by 1 μM oxytocin. (E) concentration-effect curves of oxytocin on cells expressing OXTR(T4L)-KΔ or KΔ. Values are mean +/− SEM. n ≥ 6.

The application of the natural agonist, oxytocin (1μM), triggered a large activation of the Kir6.2 channel (Fig. 4C&D), confirming the ability of the ICCR to also detect the conformational changes of this receptor in a concentration-dependent manner (Fig.4E). Comparison with the wt receptor was not possible because the wt OXTR ICCR produced no detectable signal. One reason for this could be a high basal activity of OXTR in our experimental conditions, a persistent activation of Gq-proteins and, in turn, phospholipase C which hydrolyzes the phosphatidylinositol 4,5-bisphosphate (PIP2) (Kobrinsky et al., 2000) required for the opening of Kir6.2. Other potential explanations could include an improper folding of the wt OXTR ICCR or artificial oligomeric status leading to inactivity of the channel. However it is unlikely that insertion of the T4L domain would correct the folding or the oligomeric status of the OXTR ICCR.

Discussion

These results demonstrate the ability of the ICCR technology to assess the functionality of G protein-"uncoupled" receptors that contain the insertion of the T4L domain in the i3 loop. Compared to the functional assays currently used for such engineered GPCRs (radioligand binding and fluorescence spectroscopy of bimane probes), ICCRs offer complementary advantages by reporting global conformational changes of GPCRs occurring between the ligand binding sites and the G protein interaction sites. Thus, we discovered that GPCRs(T4L) display a partial agonist phenotype (lower agonist-efficacy), probably due to the absence of the i3 loop because the same phenotype is observed for an ICCR lacking this domain. Interestingly, among the seven different GPCRs crystallized in an agonist-bound state (rhodopsin, β1 and β2 adrenergic, A2A, NTSR1, 5-HT1B and 5-HT2B receptors), five were engineered with the T4L or BRIL domain inserted in their i3 loop (β2 adrenergic, A2A, NTSR1, 5-HT1B and 5-HT2B receptors) (Venkatakrishnan et al., 2013) and seems to display an intermediate active state. A recent NMR study of β2 receptor (Kim et al., 2013) reveals, not only several active and inactive states, but also an intermediate active state with a full agonist due to the absence of interaction with a G protein or its mimic, the nanobody Nb80. This finding reinforces the interpretation that the full agonist isoproterenol generated an intermediate active state of the β2(T4L) receptor due to the absence of G protein interaction. This partial agonist effect was detected with the ICCR assays. Surprisingly, the same partial agonist effect is observed for the M2(Δi3) ICCR which is still able to interact with G proteins. This suggests that the i3 loop would be required to stabilize the receptor in the full-agonist active state.

Concentration-effect curves provided EC50s of 1.6 μM (carbachol) for M2, 141 nM (isoproterenol) for β2 and 1.5 μM (oxytocin) for OXTR(T4L) ICCRs. In the literature, values of EC50 are quite dispersed depending on the assay and the cells used. However, cAMP assays performed on mammalian cells provide similar EC50s for M2 (carbachol: 2.4 μM) (Kovacs et al., 1998) and β2 (isoproterenol: 80 nM) (Scott et al., 1999). The EC50 for OXTR determined with inositol phosphate accumulation assay in mammalian cells is 2 orders of magnitude lower (oxytocin: 10 nM) (Corbani et al., 2011).

Ligand binding to OXTR is highly dependent on cholesterol that is present in the plasma membrane of Xenopus oocytes (Hill et al., 2005). The lower apparent affinity observed in the OXTR(T4L) ICCR could be related to a partial masking of the cholesterol binding sites induced by the fusion with the channel.

By extrapolation, this technology could be relevant for the characterization or structure-function studies of GPCRs with impaired G protein binding generated by insertion of other globular soluble protein domains (Chun et al., 2012) in the cytoplasmic loops or by site-directed mutagenesis (Liu et al., 1996). Indeed, we demonstrated in this study with GPCR(T4L), and in previous studies by inhibition of Gi/o protein activation with pertussis toxin, and in cell-free recordings (excised outside-out patch-clamp mode), that the electrical signal generated by the fused Kir6.2 channel is independent of G protein activation. Based on this property, ICCR technology has the potential to detect ligand-induced conformational changes of engineered G protein uncoupled receptors.

This technology is also suitable to automation using voltage-clamp robots such as the HiClamp setup (Multichannel Systems GmbH). Assays are performed in a 96-well plate format and the characterization of one oocyte plate is completed in less than 3 hours.

This method is currently validated on class A (rhodopsin-like) GPCRs with the prerequisite of protein engineering to create functional ICCRs. However, ICCR technology has now been applied to a number of distinct receptors (Caro et al., 2011; Caro et al., 2012; Moreau et al., 2008), and simple building rules have emerged that should greatly simplify its application to a wide range of receptors.

We focused in this work on the set of GPCRs where a small domain is inserted in the intracellular loop 3. This set is certainly limited but it represents the large majority of GPCRs of known structure and is likely to expand as more and more GPCR structures are published. As demonstrated here, the method is appropriate for i3 loop-modified GPCRs; it will also be useful for any GPCR that requires complex engineering likely to affect signalling.

Experimental procedures

ICCR engineering

Ion channel-coupled receptors were created by the genetic fusion of GPCR genes to the 5' extremity of the mouse KCNJ11 gene (Kir6.2) cloned in a modified pGEMHE Xenopus oocytes vector (pGH2). For biochemical detection, a hemagglutinin tag extended by 11 residues (Schwappach et al., 2000) is present in the extracellular loop of Kir6.2. GPCR insertion was performed by a 2-step polymerase chain reactions (PCRs) using hybrid primers with complementary GPCR sequences at the 3' end and final template sequences in 5'. All PCRs were prepared with the Quikchange Lightning kit (Agilent Technologies). The first PCR amplifies the full length GPCR gene without consideration of the original vector. After in-gel DNA-purification using Geneclean turbo kit (MP Biomedicals), the PCR product serves as a "megaprimer" for a second PCR using the final template Kir6.2 pGH2. As previously reported, truncation of the first 25 residues of the channel provides the optimal functional coupling with fused GPCRs (Moreau et al., 2008). Consequently, all GPCRs were fused to the ΔN25Kir6.2 channel. The human M2 muscarinic and the β2 adrenergic ICCRs were already available at the beginning of the project and the T4L domain was inserted by the 2-step PCR described above using the β2(T4L) receptor D1 construct (Rosenbaum et al., 2007) as the original template. The last 9 residues from M2 were truncated in M2-K and M2(T4L)-KΔ generating agonist-inhibited ICCRs. In β2-K and β2(T4L)-KΔ, the last 62 residues were truncated to generate functional coupling between the receptor and the channel (Caro et al., 2011). Expression of β2 ICCRs required co-expression of the first transmembrane domain (TMD0) of the sulfonylurea receptor SUR1, a physiological partner of Kir6.2 (Caro et al., 2011; Chan et al., 2003). Concerning the human oxytocin receptor, the OXTR(T4L) gene was codon-optimized for insect cell expression, its last 42 residues were removed and the influenza A virus hemagglutinin signal sequence was inserted at its N-terminus (Liu et al., 2012).

The final DNA constructs were identified by restriction profile and verified by DNA sequencing (Beckman Coulter Genomics). After overnight linearization and standard phenol/chloroform extraction, mRNA synthesis was performed by in vitro transcription (mMessage mMachine T7 kit, Ambion) and purified by the same phenol/chloroform method. Quantitative and qualitative controls of the RNA samples were done by electrophoresis and U.V. spectrophotometry (Nanodrop2000c, Thermo Scientific).

To test the ability of M2(Δi3)-K to activate Gi/o proteins, we co-expressed the ICCR with a G protein activated channel, Kir3.4*, which was mutated to form an homotetramer (Vivaudou et al., 1997). Surface expression of the Kir3.4* channel being much higher than the ICCR, the current generated by Kir3.4* is predominant and displays clear activation of this channel by the endogenous Gi/o proteins.

Heterologous expression in Xenopus oocytes

mRNAs were diluted in RNAse-free water and micro-injected, either manually (Nanoject, Drummond), or automatically (RoboInject, Multi Channel Systems). Each oocyte received a volume of 50 nl containing the following quantity of mRNA: 4 ng of ICCRs, 2.5 ng of TMD0 and 2 ng of Kir channels. Oocytes were surgically removed from anesthetized Xenopus laevis females using procedures that conformed to European regulations for animal handling and experiments, and were approved by governmental services (Authorization N°38 08 10 granted to Michel Vivaudou by the local veterinary agency, Directeur Départemental des Services Vétérinaires, Ministère de l'Agriculture et de la Pêche, on 22 February 2008, valid until 06 July 2015) and the Institutional Ethical Committee (Ethical Committee of Commissariat à l'Energie Atomique et aux Energies Alternatives for animal experiments, assessment n°12-040 on 23 December 2012).

Oocytes were isolated by enzymatic defolliculation in 2 mg.ml−1 type 1A collagenase (Sigma-Aldrich, C9891) solution gently shaken for 2h at 19°C. Oocytes in stage V and VI were selected and stored overnight at 19°C in Barth’s solution (1 mM KCl, 0.82 mM MgSO4, 88 mM NaCl, 2.4 mM NaHCO3, 0.41 mM CaCl2, 16 mM Hepes, pH 7.4) supplemented with 100 U.ml-1 penicillin, streptomycin and gentamycin.

Microinjected oocytes were incubated individually in 96-well plates containing Barth's solution with antibiotics, for more than 2 days at 19°C.

Whole cell current recordings

Whole-cell currents were recorded using the two-electrode voltage clamp (TEVC) technique. Two setups were used: 1) a standard manual setup composed of a GeneClamp 500 amplifier (Molecular Devices), a digidata 1440A digitizer (Molecular Devices), an 8-reservoir gravity-flow perfusion system with the electrovalve Valvelink 8.2 controller (AutoMate Scientific). The voltage protocol was a succession of 5-second sweeps with alternating 500 ms steps to -50 mV, 0 mV and +50 mV from a holding potential of 0 mV (Only the values at -50 mV were used for statistics and figures). Drug application was by bath perfusion. Voltage protocols and current recordings were performed with the pClamp10.0 software (Axon); 2) an automatic setup consisting of the HiClamp robot (Multi Channel Systems). The voltage protocol was identical to that used with the manual set-up. Drug application was performed by immersion of each oocyte in continuously-stirred 200-μl reservoirs (from a 96-well plate) containing the test solutions. Membrane voltage was clamped at -50 mV and current was recorded with the dedicated software. Microelectrodes were made by pulling borosilicate capillaries (Kimble Chase n°34502 99 for manual setup or WPI n°TW150F-4 for the robot) using a P-97 micropipette puller (Sutter Instrument) and filled with filtered 3 M KCl.

Oocytes were bathed in the following high potassium solution: 91 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, 5 mM HEPES, 0.3 mM niflumic acid (a blocker of endogenous calcium-activated chloride channels), pH 7.4. The basal current, taken as the current measured in the first minute of recording, served as a rough estimate of the number of channels at the cell surface (Kir6.2 are partially open at rest in our conditions). Concentration-response experiments were obtained by subsequent applications of increasing concentrations of ligands. Fitting of Hill equation to the data was done with Origin 8 (OriginLab Corp.) without constraints on Hill coefficient and dissociation constant. All data shown represent the currents measured at -50 mV. All experiments were repeated on different oocytes from different batches at room temperature. All statistics show mean +/− SEM and statistical significance was established with unpaired two-tailed Student t-tests.

Highlights.

  • G protein-independent detection of agonist- and antagonist-bound states

  • Determination of apparent ligand affinities and efficacies on single cells

  • Recordings are suitable to automation using two-electrode voltage-clamp robots

Acknowledgments

We are grateful to D. Rosenbaum (Dallas, USA) and B. Kobilka (Stanford, USA) for the β2AR(T4L) construct, S. Seino (Chiba, Japan) for mouse Kir6.2 and K. Chan (Palo Alto, CA) for the TMD0(SUR1)-F195 construct.

This work was supported by grants to M.V. from the Agence Nationale de la Recherche (ICCR project, grant ANR-09-PIRI-0010) and the Nanosciences Foundation (NanoBioDrop project; Grenoble, France), by the National Institutes of Health grant GM089857 to V.C., by a studentship from the Region Rhone-Alpes to K.N., by a studentship from French Ministry of Research to L.N.C. The Grenoble laboratory is a member of the French National Laboratory of Excellence ≪ Ion Channel Science and Therapeutics ≫ supported by a network grant from ANR. The authors declare no conflict of interest.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Caro LN, Moreau CJ, Revilloud J, Vivaudou M. β2-Adrenergic ion-channel coupled receptors as conformational motion detectors. PloS ONE. 2011;6:e18226. doi: 10.1371/journal.pone.0018226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Caro LN, Moreau CJ, Estrada-Mondragón A, Ernst OP, Vivaudou M. Engineering of an artificial light-modulated potassium channel. PloS ONE. 2012;7:e43766. doi: 10.1371/journal.pone.0043766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Chan KW, Zhang H, Logothetis DE. N-terminal transmembrane domain of the SUR controls trafficking and gating of Kir6 channel subunits. EMBO J. 2003;22:3843. doi: 10.1093/emboj/cdg376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cherezov V, Rosenbaum DM, Hanson MA, Rasmussen SG, Thian FS, Kobilka TS, Choi HJ, Kuhn P, Weis WI, Kobilka BK, et al. High-resolution crystal structure of an engineered human beta2-adrenergic G protein-coupled receptor. Science. 2007;318:1265. doi: 10.1126/science.1150577. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chun E, Thompson AA, Liu W, Roth CB, Griffith MT, Katritch V, Kunken J, Xu F, Cherezov V, Hanson MA, et al. Fusion partner toolchest for the stabilization and crystallization of G protein-coupled receptors. Structure. 2012;20:976. doi: 10.1016/j.str.2012.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Corbani M, Trueba M, Stoev S, Murat B, Mion J, Boulay V, Guillon G, Manning M. Design, synthesis, and pharmacological characterization of fluorescent peptides for imaging human V1b vasopressin or oxytocin receptors. J Med Chem. 2011;54:2877. doi: 10.1021/jm1016208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Haga K, Kruse AC, Asada H, Yurugi-Kobayashi T, Shiroishi M, Zhang C, Weis WI, Okada T, Kobilka BK, Haga T, et al. Structure of the human M2 muscarinic acetylcholine receptor bound to an antagonist. Nature. 2012;482:551. doi: 10.1038/nature10753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hill WG, Southern NM, MacIver B, Potter E, Apodaca G, Smith CP, Zeidel ML. Isolation and characterization of the Xenopus oocyte plasma membrane: a new method for studying activity of water and solute transporters. Am. J. Physiol. Renal Physiol. 2005;289:F217–F224. doi: 10.1152/ajprenal.00022.2005. [DOI] [PubMed] [Google Scholar]
  9. Katritch V, Cherezov V, Stevens RC. Structure-function of the G protein-coupled receptor superfamily. Annu Rev Pharmacol Toxicol. 2013;53:556. doi: 10.1146/annurev-pharmtox-032112-135923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kim TH, Chung KY, Manglik A, Hansen AL, Dror RO, Mildorf TJ, Shaw DE, Kobilka BK, Prosser RS. The role of ligands on the equilibria between functional States of a g protein-coupled receptor. J Am Chem Soc. 2013;135:9474. doi: 10.1021/ja404305k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kobrinsky E, Mirshahi T, Zhang H, Jin T, Logothetis DE. Receptor-mediated hydrolysis of plasma membrane messenger PIP2 leads to K+-current desensitization. Nat Cell Biol. 2000;2:514. doi: 10.1038/35019544. [DOI] [PubMed] [Google Scholar]
  12. Kovacs I, Yamamura HI, Waite SL, Varga EV, Roeske WR. Pharmacological comparison of the cloned human and rat M2 muscarinic receptor genes expressed in the murine fibroblast (B82) cell line. J Pharmacol Exp Ther. 1998;284:507. [PubMed] [Google Scholar]
  13. Kruse AC, Hu J, Pan AC, Arlow DH, Rosenbaum DM, Rosemond E, Green HF, Liu T, Chae PS, Dror RO, et al. Structure and dynamics of the M3 muscarinic acetylcholine receptor. Nature. 2012;482:556. doi: 10.1038/nature10867. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Liu J, Blin N, Conklin BR, Wess J. Molecular mechanisms involved in muscarinic acetylcholine receptor-mediated G protein activation studied by insertion mutagenesis. J Biol Chem. 1996;271:6178. doi: 10.1074/jbc.271.11.6172. [DOI] [PubMed] [Google Scholar]
  15. Liu W, Chun E, Thompson AA, Chubukov P, Xu F, Katritch V, Han GW, Roth CB, Heitman LH, IJzerman AP, et al. Structural basis for allosteric regulation of GPCRs by sodium ions. Science. 2012;337:236. doi: 10.1126/science.1219218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Moreau CJ, Dupuis JP, Revilloud J, Arumugam K, Vivaudou M. Coupling ion channels to receptors for biomolecule sensing. Nat Nanotechnol. 2008;3:625. doi: 10.1038/nnano.2008.242. [DOI] [PubMed] [Google Scholar]
  17. Rasmussen SG, Choi HJ, Rosenbaum DM, Kobilka TS, Thian FS, Edwards PC, Burghammer M, Ratnala VR, Sanishvili R, Fischetti RF. Crystal structure of the human beta2 adrenergic G-protein-coupled receptor. Nature. 2007;450:387. doi: 10.1038/nature06325. [DOI] [PubMed] [Google Scholar]
  18. Rasmussen SG, DeVree BT, Zou Y, Kruse AC, Chung KY, Kobilka TS, Thian FS, Chae PS, Pardon E, Calinski D, et al. Crystal structure of the β2 adrenergic receptor-Gs protein complex. Nature. 2011;477:555. doi: 10.1038/nature10361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Rosenbaum DM, Cherezov V, Hanson MA, Rasmussen SG, Thian FS, Kobilka TS, Choi HJ, Yao XJ, Weis WI, Stevens RC, et al. GPCR engineering yields high-resolution structural insights into beta2-adrenergic receptor function. Science. 2007;318:1273. doi: 10.1126/science.1150609. [DOI] [PubMed] [Google Scholar]
  20. Schwappach B, Zerangue N, Jan YN, Jan LY. Molecular basis for K(ATP) aassembly: transmembrane interactions mediate association of a K+ channel with an ABC transporter. Neuron. 2000;26:167. doi: 10.1016/s0896-6273(00)81146-0. [DOI] [PubMed] [Google Scholar]
  21. Scott MG, Swan C, Jobson TM, Rees S, Hall IP. Effects of a range of beta2 adrenoceptor agonists on changes in intracellular cyclic AMP and on cyclic AMP driven gene expression in cultured human airway smooth muscle cells. Br J Pharmacol. 1999;128:729. doi: 10.1038/sj.bjp.0702829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Tan Q, Zhu Y, Li J, Chen Z, Han GW, Kufareva I, Li T, Ma L, Fenalti G, Li J, Zhang W, Xie X, Yang H, Jiang H, Cherezov V, Liu H, Stevens RC, Zhao Q, Wu B. Structure of the CCR5 chemokine receptor-HIV entry inhibitor maraviroc complex. Science. 2013 doi: 10.1126/science.1241475. published online 12 September 2013, doi: 10.1126/science.1241475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Tucker SJ, Gribble FM, Zhao C, Trapp S, Ashcroft FM. Truncation of Kir6.2 produces ATP-sensitive K+ channels in the absence of the sulphonylurea receptor. Nature. 1997;387:183. doi: 10.1038/387179a0. [DOI] [PubMed] [Google Scholar]
  24. Venkatakrishnan AJ, Deupi X, Lebon G, Tate CG, Schertler GF, Babu MM. Molecular signatures of G-protein-coupled receptors. Nature. 2013;494:194. doi: 10.1038/nature11896. [DOI] [PubMed] [Google Scholar]
  25. Vivaudou M, Chan KW, Sui JL, Jan LY, Reuveny E, Logothetis DE. Probing the G-protein regulation of GIRK1 and GIRK4, the two subunits of the KACh channel, using functional homomeric mutants. J Biol Chem. 1997;272:31560. doi: 10.1074/jbc.272.50.31553. [DOI] [PubMed] [Google Scholar]
  26. Warne T, Serrano-Vega MJ, Baker JG, Moukhametzianov R, Edwards PC, Henderson R, Leslie AG, Tate CG, Schertler GF. Structure of a beta1-adrenergic G-protein-coupled receptor. Nature. 2008;454:491. doi: 10.1038/nature07101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. White JF, Noinaj N, Shibata Y, Love J, Kloss B, Xu F, Gvozdenovic-Jeremic J, Shah P, Shiloach J, Tate CG, et al. Structure of the agonist-bound neurotensin receptor. Nature. 2012;490:513. doi: 10.1038/nature11558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Wu B, Chien EY, Mol CD, Fenalti G, Liu W, Katritch V, Abagyan R, Brooun A, Wells P, Bi FC, et al. Structures of the CXCR4 chemokine GPCR with small-molecule and cyclic peptide antagonists. Science. 2010;330:1071. doi: 10.1126/science.1194396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Zerangue N, Schwappach B, Jan YN, Jan LY. A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels. Neuron. 1999;22:548. doi: 10.1016/s0896-6273(00)80708-4. [DOI] [PubMed] [Google Scholar]

RESOURCES