GABA_B Receptor Functioning: Focus on Allosteric Modulation

Abstract

GABA_B receptors are activated by γ-aminobutyric acid (GABA), an amino acid that serves as the primary inhibitory neurotransmitter in the central nervous system, with links to a range of neurological diseases. This receptor represents a promising drug target in the field of neurological diseases for the treatment of spasticity, alcohol use disorder, anxiety, and insomnia using therapeutic options such as baclofen and phenibut. The GABA_B receptor belongs to class C of the G protein-coupled receptors and acts as an obligate heterodimer made up of two subunits, GABA_B1 and GABA_B2. Each subunit is comprised of an extracellular domain in which GABA and the above-mentioned therapeutic drugs bind and a transmembrane domain responsible for G protein activation. GABA_B features a unique allosteric mechanism for signal transduction in which agonist binding in the GABA_B1 extracellular domain elicits G protein activation via rearrangement of the intracellular face of the GABA_B2 transmembrane domain. More recently, the GABA_B receptor has been found to display a tendency to form oligomers, thus increasing the complexity of allosteric communications in the receptor. This chapter summarizes the current state-of-the-art relating to the structure, activation mechanism, and allosteric modulation of the GABA_B receptor, which may provide novel opportunities for the development of approaches aimed at regulating receptor activity.

1 Introduction

Since their discovery in the mid-1980s as the molecular target of the anti-spasticity drug baclofen, γ-aminobutyric acid (GABA) type-B (GABA_B) receptors have garnered considerable interest, with studies focused on elucidating their molecular bases (Bowery et al. 2002; Gassmann and Bettler 2012; Pin and Bettler 2016). Subsequently, toward the end of the 1990s, a discovery was made revealing the presence of two genes encoding GABA_B receptors. The first gene encodes two variants, GABA_B1a and GABA_B1b, via an alternative initiation site, adding two sushi domains (SDs) at the N-terminus of the GABA_B1a variants (Kaupmann et al. 1997). The second subunit, GABA_B2, is structurally homologous to GABA_B1 and is essential in achieving agonist high affinity and effective G-protein coupling of the GABA_B receptor heterodimer (Jones et al. 1998; Kaupmann et al. 1998; White et al. 1998). The GABA_B2 subunit has likewise proved to be crucial for proper membrane insertion of the GABA_B receptor. Indeed, when expressed alone, GABA_B1 subunit remains intracellularly retained between the endoplasmic reticulum and Golgi due to an intracellular retention signal in its C-terminal tail (Margeta-Mitrovic et al. 2000; Pagano et al. 2001). However, interaction with the GABA_B2 subunit masks the retention signal and facilitates transportation of the heterodimer to the cell surface, thus highlighting the unique nature of this receptor in the context of G protein-coupled receptors (GPCRs) known at the time and representing the first obligatory heterodimeric GPCR. This mechanism of retention and trafficking of the GABA_B receptor seems to be highly conserved during evolution, as recently demonstrated with the drosophila GABA_B receptor (Zhang et al. 2020). Moreover, early studies demonstrated the implication of the GABA_B1 subunit in agonist binding, with GABA_B2 subunit proving essential in G protein activation (Galvez et al. 2001; Margeta-Mitrovic et al. 2001a, b). These findings represented a major breakthrough not only in the field of GABA_B receptor studies but also in the wider GPCR community, in which the notion of GPCR dimerization was still the subject of intense debate (Sleno and Hébert 2019).

The general organization of the GABA_B receptor is similar to that of other obligate dimeric class C GPCRs (Kniazeff et al. 2011), including receptors activated by glutamate (Koehl et al. 2019; Lin et al. 2021; Seven et al. 2021), calcium (He et al. 2024; Zuo et al. 2024) and sweet, and umami-tasting compounds (Xu et al. 2004) (Fig. 2.1a, b). The extensive GABA_B1 and GABA_B2 extracellular domains comprise a Venus flytrap-like (VFT) domain similar to the binding domain of metabotropic glutamate (mGlu) receptors and other class C GPCRs, with the VFT domain linked in both subunits to a heptahelical transmembrane (7TM) domain common to all GPCRs through a 10–15 amino-acid stalk region (Mao et al. 2020; Papasergi-Scott et al. 2020; Park et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021). The GABA binding site (Galvez et al. 1999) is only present in the VFT domain of the GABA_B1 subunit, whereas the GABA_B2 subunit alone is responsible for G protein coupling through rearrangement of the intracellular face of its transmembrane domain. More recent studies reported how the GABA_B receptor displays a tendency to form oligomers, thus increasing the complexity of allosteric communications in the receptor (Maurel et al. 2008; Comps-Agrar et al. 2011; Stewart et al. 2018; Xue et al. 2019); the formation of spontaneous, stable, and well-organized oligomers is a characteristic exclusive to the GABA_B receptor in the GPCR family (Pin et al. 2009).

This unique organization of the GABA_B receptor has prompted the undertaking of extensive studies aimed at elucidating the molecular bases of the activation mechanism and allosteric properties of this receptor. The atomic structures of the full-length GABA_B receptor, including its complex with G protein (Shen et al. 2021), were recently revealed by five independent groups using cryo-electron microscopy (cryo-EM) (Mao et al. 2020; Papasergi-Scott et al. 2020; Park et al. 2020; Shaye et al. 2020, 2021; Kim et al. 2020) (Fig. 2.1a, b). These studies revealed a mode of coupling of the G protein that differed from those of other GPCRs, although it was conserved in other class C GPCRs (Lin et al. 2021; Seven et al. 2021; He et al. 2024; Zuo et al. 2024). These structures also contributed toward clarifying the mode of action of positive allosteric modulators (PAMs), all of which were found to bind in the transmembrane interface formed by the GABA_B1 and GABA_B2 subunits (Mao et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021), thus at variance with the known binding sites of allosteric modulators of the class C GPCRs (Kumar et al. 2018; Seven et al. 2021; He et al. 2024; Zuo et al. 2024) and most other classes of GPCRs (Thal et al. 2018). These atomic structures will contribute to the development of novel therapeutic applications, including novel allosteric modulators aimed at modulating GABA_B receptor activity.

In this chapter, we discuss the current state of the art relating to the structure and molecular basis of GABA_B receptor activation to enhance our understanding of how this multidomain membrane protein is activated by a small ligand for the control of G protein activity. This chapter will address the issue of how allosteric transitions between the different domains regulate receptor activity and how the latter is controlled by allosteric modulators. A deeper understanding of these issues will undoubtedly provide novel options to be applied in the development of innovative treatments for use in regulating this important brain receptor and may hopefully shed light on possible assembly and allosteric interactions between other GPCRs.

2 Structure and Mechanism of Activation of the GABA_B Receptor

Resolution of the crystal structures of the isolated extracellular domains of GABA_B1 and GABA_B2 subunits (Geng et al. 2012, 2013) and, more recently, of the full-length receptor (Mao et al. 2020; Papasergi-Scott et al. 2020; Park et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021) have provided major insight into the molecular mechanism of activation of GABA_B receptors. In the absence of ligand (apo state), GABA_B1 and GABA_B2 VFT domains share a good structural homology (Geng et al. 2013). These VFT domains derive from bacterial periplasmic amino acid-binding proteins such as the leucine/isoleucine/valine binding protein (LIVP) (O’Hara et al. 1993). Each VFT domain (approx. 410 residues) (Rondard et al. 2011) is composed of two lobes linked by three short loops, with lobe 1 being the N-terminal lobe and lobe 2 the C-terminal one (Fig. 2.1a, b). The GABA_B1 VFT domain was observed in two separate conformations, being open in apo and antagonist-bound states and closed in the presence of agonists (Geng et al. 2012, 2013). In comparison, GABA_B2 VFT domain is invariably observed in an open conformation, with the angle defined by the two lobes remaining virtually constant for GABA_B2 VFT domain in all structures. The angle defined by the two lobes in GABA_B2 VFT domain is in line with the large angle observed for the apo state and antagonist-bound conformations of GABA_B1 VFT domain. The GABA_B2 VFT domain, however, remains in an open conformation, both alone and when associated with the open or closed GABA_B1 VFT domain (Geng et al. 2012, 2013). This confirms the observation whereby GABA_B1 subunit is the sole subunit binding agonist in the GABA_B receptor, underlying activation of the entire receptor complex (Kniazeff et al. 2002, 2004).

Fig. 2.1

GABA_B receptor structural organization. (a, b) Cryo-electronic map of the baclofen bound to GABA_B receptor in complex with the positive allosteric modulator, rac-BHFF, and the Gi protein (PDB 7EB2) (a) and its model ribbon view (b), respectively. (c) Positive allosteric interactions between GABA_B1 and GABA_B2 a Venus flytrap-like (VFT) domains (Liu et al. 2004) and between GABA_B1 and GABA_B2 heptahelical transmembrane (7TM) domains (Monnier et al. 2011) are illustrated

Structural and mutagenesis studies have been performed, describing the GABA binding site in GABA_B1 VFT domain in detail (Galvez et al. 1999; Galvez et al. 2000; Kniazeff et al. 2002; Geng et al. 2013). The carboxylate moiety of GABA lies at the center of a hydrogen-bound network involving Ser 246 and Ser 269 in lobe 1 and Tyr 366 in lobe 2 (in the whole chapter, indicated residues correspond to GABA_B1a subunit numbering). The γ-amino group interacts with His 286 and Glu 465 in lobe 1 and with Trp 394 in lobe 2 through hydrogen-bound and van der Waals contacts. Baclofen, a GABA_B receptor-specific agonist, binds in a similar way to GABA but with a conformational flip of Tyr 366 to accommodate the chlorophenyl moiety of the ligand (Galvez et al. 2000; Geng et al. 2013). Orthosteric receptor antagonists are GABA derivatives that also bind to GABA_B1 VFT domain alone. Co-crystallization of GABA_B VFT domains with antagonists showed how they bind tightly to lobe 1, involving similar residues to GABA binding (Ser 246, Ser 269, His 286, Glu 465, and Trp 181). However, compared to agonist-bound conformation, only sparse interaction with lobe 2 is detected, in line with the greater distance between the two lobes and stabilization of an “open” conformation (Geng et al. 2013). Of note, the residues involved in GABA binding in GABA_B1 VFT domain are not conserved in GABA_B2 VFT domain (Kniazeff et al. 2002). In addition, in contrast to the GABA_B1 VFT domain cleft where agonists bind, the GABA_B2 cleft displays no specific or high conservation during evolution, strongly suggesting the absence of ligand interaction at this site (Kniazeff et al. 2002).

Upon agonist binding in GABA_B1 VFT domain, the VFT dimer undergoes relative rearrangement. The lobe 1 interface, situated between the two subunits consisting of a central hydrophobic patch surrounded by hydrogen bonds and a salt bridge, serves as a rotation axis. As a consequence, a higher proximity and interaction between the two lobes 2 is observed in agonist-bound structures, with additional contacts between the lobes 2 involving mainly polar interactions. Of note, this relative reorientation between the two VFT domains in the GABA_B receptor is common to class C GPCRs containing a VFT domain such as that observed in mGlu receptors and calcium sensing (CaS) receptors, although the amplitude of this movement is considerably smaller in GABA_B receptors. Our group has taken advantage of this reorientation of GABA_B to develop fluorescence resonance energy transfer (FRET)-based conformational sensors (Lecat-Guillet et al. 2017) aimed at recording agonist-induced activation of the receptor by means of light measurement in high-throughput assays. However, consistent with the low amplitude reorientation of GABA_B VFT domains, the position of the fluorophores was optimized to observe a change of FRET signal upon agonist activation in these GABA_B receptor conformational sensors (Lecat-Guillet et al. 2017), compared to other sensors developed for mGlu receptors (Doumazane et al. 2013; Scholler et al. 2017; Lecat-Guillet et al. 2023) and CaS receptors (Liu et al. 2020). Indeed, in these latter conformational sensors, the attachment of fluorophores at the N-terminal end of the mGlu and CaS receptors was sufficient to measure a large change in the FRET signal, indicating a major conformational change in the reorientation of the VFT dimer in these receptors.

The cryo-EM structure of the full-length GABA_B receptor also revealed the first atomic structures of GABA_B 7TM domains (Fig. 2.1a, b), homologous to those found in all other GPCRs, including other class C receptors. However, the conformational change observed between the inactive and active states is more subtle, with a smaller movement of the transmembrane helix 6 (TM6) compared to class A and B GPCRs. Indeed, movement of the TM6 linked to receptor activation and G protein coupling represents one of the main characteristics of the activation mechanism of GPCRs (Weis and Kobilka 2018). A similarly low amplitude of this TM6 movement is conserved in other class C GPCRs, including the mGlu and CaS receptors, displaying consistency with a different mode of coupling of the G protein to the GABA_B receptor (Shen et al. 2021). Indeed, surprisingly, the structure of the GABA_B receptor in complex with the G protein has revealed how the latter interacts at a different position to the intracellular face of the class C receptors. In addition, as a consequence of the subtle movement of TM6, the C-terminal end of the Gα subunit does not penetrate deeply into the 7TM core of the GABA_B2 subunit but engages mainly with intracellular loops i2 and i3, as found in other class C mGlu (Lin et al. 2021; Seven et al. 2021) and CaS (He et al. 2024; Zuo et al. 2024) receptors. It is therefore rather a challenge to grasp how the GABA_B2 subunit activates the G protein by inducing the release of guanosine 5′-diphosphate (GDP) bound to the Gα subunit.

Another important feature is constituted by the 7TM heterodimer interface mediated by the two transmembrane helices 5 (TM5s) of GABA_B1 and GABA_B2 subunits in the inactive state, and TM6s in the active state (Mao et al. 2020; Papasergi-Scott et al. 2020; Park et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021). This relative orientation of the 7TM domains is in agreement with a recent study conducted in our lab using site-directed mutagenesis and structure-function analysis, in which the interface of dimerization was revealed by means of a cysteine crosslinking approach (Xue et al. 2019). This study showed how it is sufficient to stabilize the interaction between the TM6s of GABA_B1 and GABA_B2 subunits to induce constitutive activity of the receptor. Of note, the subtle rearrangement of the 7TM interface in GABA_B receptors between the inactive and active states compared to the mGlu receptors (Xue et al. 2015, 2019; Seven et al. 2021; He et al. 2024; Krishna Kumar et al. 2024; Zuo et al. 2024), is consistent with the subtle rearrangement described above in the VFT heterodimer between the inactive and active states. Finally, an important feature revealed by GABA_B receptor structures is the stalk region that maintains a space between the VFT and 7TM domains in each subunit (Mao et al. 2020; Papasergi-Scott et al. 2020; Park et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021). The long extracellular loop 2 of each subunit protrudes from the 7TM core and interacts directly with these stalk regions. These interactions are most probably crucial in transmitting activation from VFT to 7TM domains, although the mechanism underlying this coupling remains unclear.

Unfortunately, the C-terminal regions of the GABA_B receptor were not observed, as truncated subunits were used, and also due to the high flexibility of these regions. Indeed, the intracellular C-terminal region of GABA_B1 and GABA_B2 subunits is rather long (107 and 200 residues, respectively) and contains a well-structured coiled-coil domain essential in achieving heterodimerization and ensuring a correct assembly of the heterodimer prior to appropriate targeting of the plasma membrane (Kammerer et al. 1999; Margeta-Mitrovic et al. 2000; Pagano et al. 2001). Heterodimeric interaction of the GABA_B receptor is stabilized by a coiled-coil interaction between the GABA_B1 and GABA_B2 C-termini, encompassing approximately 49 residues in each subunit (Ser 772–His 810 in GABA_B1 and Ser 779 Lys 827 in GABA_B2) (Kammerer et al. 1999; Burmakina et al. 2014). The structure of GABA_B receptor coiled-coil domain has been solved by X-ray crystallography, highlighting molecular details of the interaction (Burmakina et al. 2014). The structure of this domain is conserved during evolution, as recently shown by the structure of the coiled-coil domain of the drosophila GABA_B receptor by nuclear magnetic resonance (NMR) (Zhang et al. 2020).

Finally, an additional structural domain is present on one of the isoforms of the GABA_B receptor. Indeed, two main isoforms, GABA_B1a and GABA_B1b, of the GABA_B1 subunit are generated through an alternate promoter usage (Steiger et al. 2004), resulting in the presence of a repeat of two sushi domains (SDs, named SD1 and SD2) at the extracellular N-terminus of GABA_B1a subunit alone. SDs, also known as complement control protein (CCP) modules or short consensus repeats (SCR), are approx. 60 residues long and known to be involved in numerous recognition processes, including that of the complement system (Reid and Day 1989). In the case of the GABA_B receptor, SDs control specific targeting of the receptor to excitatory terminals, most probably through interactions with the extracellular matrix (Vigot et al. 2006). SD2 is a typical SD having approximately 60 amino acid residues, including four cysteines forming two disulfide bridges (Blein et al. 2004). NMR analysis revealed the three-dimensional structure of SD2 to be similar to that of previously solved SD structures. In contrast to SD2, SD1 has less sequence homology with typical SDs and is unstable when purified alone or when fused to SD2 (Blein et al. 2004).

3 Allosteric Interaction Within the GABA_B Receptor and with the G Protein

The following section illustrates how allosteric communication between the four main domains in GABA_B receptor concurs toward activation of the receptor (Fig. 2.1c). First, a positive allosteric interaction between GABA_B1 and GABA_B2 VFT domains was demonstrated (Liu et al. 2004; Geng et al. 2012), although GABA_B2 VFT domains also play an important role in facilitating GABA_B1 VFT closure, thereby resulting in a higher affinity for orthosteric receptor agonists (Liu et al. 2004; Geng et al. 2012). The tighter interaction between GABA_B1 and GABA_B2 VFTs in the presence of an agonist stabilizes the closed state of GABA_B1 VFT domain, increasing agonist potency and revealing a positive allosteric interaction between the two VFT domains (Fig. 2.1c). However, while interactions between the lobes 2 are strictly required for activation of the wild-type receptor, a GABA_B receptor mutant lacking GABA_B2 VFT domain is still functional, although displaying low agonist potency and low efficacy (Monnier et al. 2011).

A second major positive allosteric modulation in the GABA_B receptor is represented by functional interaction between the two 7TM domains (Monnier et al. 2011) (Fig. 2.1c). Indeed, agonist-induced activity of the GABA_B mutant lacking the GABA_B2 VFT domain described above has demonstrated the evidence of a strong allosteric coupling between the two 7TMs, referred to as transactivation. In this mechanism, coupling between the agonist-bound VFT and 7TM domains in GABA_B1 subunit is sufficient to stabilize the GABA_B2 7TM in the active state through the 7TM dimer interface. The latter also highlights the importance of reorientation of the heterodimer interface from TM5-TM5 to TM6-TM6 interface in receptor activation, a conformational change expected to occur prior to activation of GABA_B2 7TM core and G protein coupling (Shaye et al. 2020), according to the molecular mechanism of activation proposed for the mGlu receptors (Rondard and Pin 2015). It however remains unclear how this conformational change in the VFT dimer leads to activation of the GABA_B2 7TM domain, although most likely implicating stalk domains and an interaction of these with the extracellular loops 2 of the two subunits.

The G protein is known to exert a positive allosteric effect in the context of stabilizing the active conformation of the 7TM domain of GPCRs (DeVree et al. 2016). The latter was well demonstrated using a FRET-based conformation sensor of GABA_B receptor that reports conformational states of the VFT dimer (Lecat-Guillet et al. 2017). G protein mutants that facilitate coupling of the G protein to the GABA_B receptor favor rearrangement of the VFT dimer to the active state upon agonist activation, as previously observed for the mGlu₂ receptor (Doumazane et al. 2013). Interestingly, using bioluminescence resonance energy transfer (BRET)-based G protein sensors, we have recently shown that native GABA_B receptors in neurons couple to all Gi/o proteins in the same manner observed in human embryonic kidney 293 (HEK 293) cells, although agonist-induced activation of the G proteins by native receptors in neurons proved to be much more efficient (Xu et al. 2024). These studies also revealed how the native GABA_B receptors in neurons do not seem to possess constitutive activity, in contrast to the recombinant receptors expressed in cells (Xu et al. 2024; Ma et al. 2024). This finding might be explained by the fact that GABA_B receptors expressed in neurons are capable of interacting with both extracellular and intracellular proteins, likely implicated in the control of the conformational state and limitation of activation.

4 Allosteric Modulation by Positive Allosteric Modulators

In addition to the orthosteric receptor agonists and antagonists, allosteric modulators were developed to control the activity of the GABA_B receptor by acting on other binding sites in the receptor (Urwyler 2011) (Fig. 2.2a). The GABA_B PAMs are of special interest since they enable to favor the active conformation of the receptor. They are also expected to respect the physiological activity of the receptor by potentiating the activation by GABA, without having an agonist activity per se (Pin and Prézeau 2007), so that they may have better therapeutic efficacy, with fewer side effects.

Fig. 2.2

Positive allosteric modulation of the GABA_B receptor. (a) Chemical structure of the main positive allosteric modulators (PAMs) of GABA_B receptors is indicated. CGP7930 (Urwyler et al. 2001), rac-BHFF (Malherbe et al. 2008), GS39783 (Dupuis et al. 2006), BHF177 (Guery et al. 2007), ADX71441 (Kalinichev et al. 2017), COR627, and COR628 (Castelli et al. 2012). (b) Binding site of the indicated GABA_B PAMs at the interface between the heptahelical transmembrane (7TM) domains and, notably, transmembrane helices 6 (TM6s) of GABA_B1 and GABA_B2 was validated by site-directed mutagenesis and functional analysis (Liu et al. 2021)

Confirmation of the molecular basis underlying the action of these PAMs has recently been clarified using GABA_B receptor structures (Mao et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021) together with FRET-based conformational GABA_B sensors (Lecat-Guillet et al. 2017) and BRET-based G protein sensors (Xu et al. 2024). Indeed, while these compounds were expected to bind to the GABA_B2 7TM core (Binet et al. 2004; Freyd et al. 2017), surprisingly, the solving of structures revealed how they all bind at the 7TM heterodimer interface, stabilizing the active interface formed by the two TM6 helices (Mao et al. 2020; Shaye et al. 2020; Kim et al. 2020; Shen et al. 2021) (Fig. 2.2b). In addition, the binding site of three of these compounds, CGP7930, GS39783 (Shaye et al. 2020) and rac-BHFF (Shen et al. 2021) has been validated by site-directed mutagenesis (Liu et al. 2021). These GABA_B PAMs bind in the same binding pocket located deep inside the transmembrane domain interface, close to the intracellular part of the receptor. This unexpected PAM-binding site might be explained by difficulties encountered in binding to the 7TM core, occupied by phospholipids in the inactive state (Papasergi-Scott et al. 2020; Park et al. 2020; Kim et al. 2020). This is in contrast to PAMs for class C GPCRs, which interact in the 7TM core, with the exception of one PAM that binds at the interface of the heterodimeric mGlu_2–4 receptor in the vicinity of the extracellular face (Wang et al. 2023).

The intrinsic agonist activity of these compounds has likewise been recently clarified. In transfected cells, CGP7930, and largely rac-BHFF, appear to exert intrinsic agonist activity in the absence of GABA (Lecat-Guillet et al. 2017; Liu et al. 2021), while, conversely, GS39783 is a pure GABA_B PAM (Lecat-Guillet et al. 2017). As expected, other compounds such as COR627 and COR628 also behaves as a pure GABA_B PAM (Castelli et al. 2012). Of interest, agonist activity displayed by rac-BHFF was not observed for the native GABA_B receptor in neurons, indicating that developed PAMs might act as pure allosteric modulators in the native system (Xu et al. 2024).

In contrast, negative allosteric modulators (NAMs) of GABA_B receptors were poorly developed compared to potentiators, while competitive receptor antagonists were reported to improve cognition in animal models (Froestl et al. 2004; Kleschevnikov et al. 2012; Iqbal and Gillani 2016). So far, only two series of GABA_B NAMs derived from CGP7930 (Sun et al. 2016) and COR compounds (Porcu et al. 2021) have been developed but not tested in vivo.

5 Allosteric Modulation Through Oligomerization

The GABA_B receptor has been reported to display a tendency to form organized and stable oligomers at the cell surface (Maurel et al. 2008; Comps-Agrar et al. 2011, 2012; Calebiro et al. 2013), an unexpected property for GPCRs outside rhodopsin receptors in the retina (Jastrzebska 2017). The presence of tetramers and higher-order oligomers has also been proposed based on the findings of FRET (Maurel et al. 2008) and diffusion (Calebiro et al. 2013) studies of the GABA_B receptor performed in transiently transfected mammalian cells expressing even very low expression levels of receptor, compatible with endogenous expression in cortical neurons (Maurel et al. 2008; Comps-Agrar et al. 2011).

Several experiments have demonstrated how the association of GABA_B receptor heterodimers is mediated by GABA_B1 subunits (Fig. 2.3a). Primarily, in recombinant transfected HEK 293 cells, a high FRET signal is measured between the SNAP-tagged GABA_B1 subunits labeled with fluorophores compatible with FRET measurement, while this signal is low between the SNAP-tagged GABA_B2 subunits. Accordingly, a significant FRET signal was measured for the endogenous GABA_B receptor in mouse brain membranes using fluorescent anti-SD antibodies expected to recognize GABA_B1a subunits (Tiao et al. 2008; Comps-Agrar et al. 2011).

Fig. 2.3

Allosteric modulation in GABA_B receptor oligomers. (a) Schemes for the organization of transmembrane domains in the GABA_B oligomer where the GABA_B1 subunits form the interface between the heterodimers in the inactive and active states, as recently proposed (Maurel et al. 2008; Comps-Agrar et al. 2011; Stewart et al. 2018; Xue et al. 2019). (b) Model for the organization of the heptahelical transmembrane (7TM) domains in GABA_B oligomers in the inactive and active states based on a cysteine-crosslinking analysis (Xue et al. 2019). In this study, the close proximity between transmembrane helices at the interface between GABA_B1 and GABA_B2 7TM domains, and between two GABA_B1 subunits was mapped in absence or presence of GABA. GABA_B2 subunits are shown in transparency. (c), The negative allosteric interactions between GABA_B heterodimers limit the efficacy of G protein activation, as demonstrated (Maurel et al. 2008; Comps-Agrar et al. 2011; Stewart et al. 2018)

The proximity between GABA_B1 VFT domains is also consistent with cysteine crosslinking studies of the GABA_B1 7TM domain (Fig. 2.3b). Indeed, specific disulfide crosslinkings were obtained between TM5s in the active state that were further increased following agonist treatment, indicating that GABA_B1 TM5s may be located in close proximity. Interestingly, in this study (Xue et al. 2019), our group mapped the interface of the interaction between GABA_B heterodimers in higher-order oligomers and revealed the conformational changes of these interfaces upon receptor activation (Fig. 2.3b).

Nevertheless, the functional consequences of GABA_B oligomerization on receptor function should be ascertained. To determine the differential G protein coupling profiles of heterodimers and oligomers, our group has developed a series of strategies aimed at decreasing the level of receptor oligomerization in cells. As a first step, a minimal construct consisting of the 7TM part of GABA_B1 subunit but lacking the VFT domain and the C-terminal region was used as a competitor of the GABA_B1/GABA_B1 interface (Maurel et al. 2008; Comps-Agrar et al. 2011). Subsequently, we engineered mutations in GABA_B1 VFT domain to produce an impairment at the GABA_B1/GABA_B1 VFT interface based on the structure of VFT tetramers in the glutamate receptor, AMPA, involving an interaction with VFT lobe 2 (Sobolevsky et al. 2009; Comps-Agrar et al. 2011; Stewart et al. 2018). Both strategies yielded a GABA_B receptor featuring a higher G protein coupling efficacy without affecting the potency of GABA stimulation. This highlights the critical role played by oligomerization in controlling G protein coupling efficacy, further underlining the superior efficiency of oligomers in limiting G protein coupling compared to heterodimers.

Binding experiments performed on GABA_B1 subunit mutants, in which oligomerization was reduced by the introduction of an N-glycosylation for the purpose of inducing steric hindrance (Stewart et al. 2018), allowed us to propose a model of allosteric interactions with the GABA_B heterodimer in these oligomers (Fig. 2.3c). The GABA_B1 subunit of one heterodimer was able to interact with a second GABA_B1 subunit to limit ligand binding. In addition, one of these N-glycosylation mutants resulted in an increase of G protein efficacy. This further highlights the critical role played by oligomerization in controlling G protein coupling efficacy.

6 Conclusion

Recent structural, biophysical, and biochemical studies carried out to study the GABA_B receptor have highlighted the high degree of complexity of this receptor. The receptor indeed appears to be well arranged in organized oligomers aimed at controlling G protein signaling. This organization enables to strictly control allosteric interactions between heterodimers and between domains within heterodimer, which, by all accounts, are implicated in the activation and modulation of the receptor. Finally, the complex structure of the GABA_B receptor is well suited to foster the development of novel allosteric modulators. However, whether binding to the heterodimer interface constitutes the only means of developing these allosteric modulators remains an open question to be discussed in the context of the development of future drugs.