Minireview: Role of Intracellular Scaffolding Proteins in the Regulation of Endocrine G Protein-Coupled Receptor Signaling

Abstract

The majority of hormones stimulates and mediates their signal transduction via G protein-coupled receptors (GPCRs). The signal is transmitted into the cell due to the association of the GPCRs with heterotrimeric G proteins, which in turn activates an extensive array of signaling pathways to regulate cell physiology. However, GPCRs also function as scaffolds for the recruitment of a variety of cytoplasmic protein-interacting proteins that bind to both the intracellular face and protein interaction motifs encoded by GPCRs. The structural scaffolding of these proteins allows GPCRs to recruit large functional complexes that serve to modulate both G protein-dependent and -independent cellular signaling pathways and modulate GPCR intracellular trafficking. This review focuses on GPCR interacting PSD95-disc large-zona occludens domain containing scaffolds in the regulation of endocrine receptor signaling as well as their potential role as therapeutic targets for the treatment of endocrinopathies.

Endocrine function is regulated by a myriad of hormones with the majority signaling their targets through G protein-coupled receptors (GPCRs). In humans, the superfamily GPCRs represents the largest family of cell surface proteins with approximately 800 GPCR genes identified to date (1–3). Even though the members of the GPCR family share a common molecular architecture consisting of seven-transmembrane domains connected by three extracellular and three intracellular loops, they display significant sequence and functional diversity (4–6). GPCRs are the target of a vast variety of ligands and thus function as molecular machines and transmit an incredible wide range of extracellular signals across the cell membrane as intracellular signals that are recognized by the cell. GPCRs are involved in the regulation of virtually all essential physiological processes in the human body. Thus, it is easy to appreciate that the dysfunction of GPCR-mediated signaling pathways can cause pathological states (7). Although GPCRs comprise only approximately 3% of all the genes encoded by the human genome, they represent the target for 40%–50% of all prescription drugs (3, 8), and endocrine GPCRs play a pivotal role as sensors regulating endocrine function (9).

GPCRs act as ligand-regulated guanine nucleotide exchange factors for heterotrimeric GTP-binding proteins (G proteins) and thereby transduce extracellular signals into intracellular second messengers. Upon ligand binding GPCRs undergo a conformational change to adopt an active conformation. This facilitates the exchange of GDP for GTP on the G protein Gα-subunit and leads to the dissociation of the Gα-subunit from the receptor and the Gβγ-subunit. The G protein subunits then regulate a variety of downstream effectors, such as ion channels, phospholipases, and adenylyl cyclases (10). However, over the past 2 decades, new concepts for GPCR signal transduction have been described making the regulation of GPCR signaling more diverse and complicated than previously anticipated. In addition to the classical paradigm of G protein-mediated signal transduction, GPCRs have been shown to signal via G protein-independent mechanisms such as β-arrestin-mediated signaling pathways (2, 11–14). This has revealed that heterotrimeric G proteins are not the only GPCR interacting proteins (GIPs) that modulate GPCR signal transduction. It is now appreciated that the nature, kinetics, and strength of GPCR signaling is modulated by a vastly growing number of GIPs (see reviews in references 15–19). GPCR scaffolded proteins are capable of associating with multiple proteins that enable the fine-tuning of GPCR signaling (20). Consequently, the GPCR-dependent scaffolding of intracellular regulatory proteins provides the potential for the development of novel strategies that target specific intracellular protein interactions with GPCR that regulate specific GPCR intracellular signaling pathways. Specifically, this type of strategy may be of therapeutic value for the treatment of endocrine-related diseases. This review will focus on the current, yet evolving, understanding of the regulation of endocrine signaling by GPCR-scaffolded proteins and the potential targeting GPCR-scaffolded complexes for the treatment of endocrine diseases.

GPCR signal transduction

Upon agonist stimulation, in addition to heterotrimeric G protein coupling, the intracellular domains of the GPCR serves as substrates for phosphorylation by either second messenger-dependent protein kinases or G protein-coupled receptor kinases (GRKs) (2, 11–14). GPCR agonist activation and GRK phosphorylation subsequently results in the recruitment of β-arrestin proteins that contribute to the waning of GPCR responsiveness to repeated agonist stimulation; an important physiological mechanism that is termed desensitization. In addition to terminating G protein signaling, β-arrestin proteins also function as adaptor proteins that target GPCRs for clathrin-mediated endocytosis via their ability to recruit both clathrin and β-adaptin (11, 21–26). After internalization various different subcellular trafficking events take place to either restore GPCR responsiveness or to target the down-regulation of GPCRs (11, 27).

The activation of a GPCR due to agonist binding leads to a conformational change in the receptor (transmembrane helix rearrangement) (28), which promotes the direct interaction of the GPCR with heterotrimeric G proteins via the four intracellular domains. Heterotrimeric G proteins are composed of α-, β-, and γ-subunits (29). There are 16 known α-subunits, which are divided into four subfamilies based on sequence and functional similarities: Gα_S, Gα_i, Gα_q, and Gα_12/13. G_i subfamily members decrease the level of the intracellular second messenger cAMP by inhibiting adenylyl cyclase activity or activating phosphodiesterases. Members of the G_s family activate adenylyl cyclase, which results in the production of cAMP and the subsequent stimulation of protein kinase A (PKA), a process that is antagonized by Gα_i-coupled GPCRs. The activation of Gα_q proteins leads to the activation of phospholipase C (PLC), the formation of inositol 1,4,5 trisphosphate (IP3) and diacylglycerol, the release of intracellular Ca²⁺ stores, and the subsequent activation of protein kinase C (PKC). Members of the G_12/13 family are well known to regulate small GTPases like Rho (30–32). GPCR/G protein binding leads to the activation of G proteins due to the release of GDP and binding of GTP to the Gα subunit (20, 33). The Gα- and Gβγ-subunits then dissociate and mediate the regulation of the various downstream effector systems mentioned above which results in changes in cellular physiology.

This pathway, comprising monomeric/dimeric GPCRs, heterotrimeric G proteins, and multiple effector systems, is referred to as the canonical signaling pathway. However, this basic concept is somewhat outdated and has been revised over the past decades of GPCR research. We now know that there are various other GIPs other than G proteins that enable fine-tuning of signaling in a receptor subtype- and tissue-specific manner (5). GIPs play crucial roles in receptor signaling as they can do one of the following: 1) directly mediate receptor signaling, 2) control receptor localization and/or trafficking, 3) act as allosteric modulators, or 4) act as scaffold proteins that link the receptor to downstream effectors (20). This has led to the discovery of potential mechanisms for GPCR signaling that extend beyond classical paradigms. Because GPCRs can interact with multiple G protein and non-G protein effectors, it is now appreciated that GPCR signaling is pluridimensional (34).

In past years, several receptors have been identified that are either able to couple to multiple G proteins or differentially activate G proteins different tissues thereby enable a single GPCR to engage in several signaling pathways simultaneously, ie, muscarinic acetylcholine receptors, human TSH receptor (TSHR), and D1 dopamine receptor (35–38). However, G protein signaling alone does not qualify for the term pluridimensional GPCR signaling. It was the discovery of the underlying robust G protein-independent signaling network that revealed the true GPCR efficacy (13, 37, 39). The key players in G protein-independent signaling pathways and most likely the best characterized non-G protein effectors are the arrestins. This family of cytosolic scaffolding proteins is well known for its critical role in GPCR desensitization. As indicated above, β-arrestin binding not only precludes G protein activation and targets GPCRs for endocytosis, the nonvisual arrestins also initiate a second wave of signaling that takes place as soon as G protein signaling vanishes and as receptors redistribute to intracellular compartments (39). Less than 20 years ago, it was believed that β-arrestins' exclusive role in GPCR signaling is binding to the receptor to arrest G protein-mediated signaling (11). However, in the original studies demonstrating additional roles for β-arrestin function showed that they also scaffold protein complexes required for the initiation of endocytosis (clathrin and β-adaptin) and signaling (c-Src) (13, 21, 23, 25). The first example of β-arrestin signaling was reported in 1999 by Luttrell et al (13), who discovered β-arrestin-dependent formation of a c-Src signaling complex with the β₂-adrenergic receptor (β₂AR). Furthermore, recent evidence suggests that not only GPCR desensitization but also internalization is a signaling event (11, 24). Particularly, the introduction of new techniques, such as conformational biosensors (40), has generated the hypothesis that cell signaling events occur within intracellular compartments (41). Such signaling processes from endosomes either include dynamic sites of GPCR/G protein activation or comprise signal control in endosomes independent of G proteins (reviewed in references 41–43).

Calebiro et al (44) and Ferrandon et al (45) discovered the requirement of GPCR/G protein internalization for sustained or persistent Gα_s signaling from an endosomal compartment. Although the TSHR was found in a pre-Golgi compartment in close association with the Gα_s subunit and adenylyl cyclase III (44), the internalization of the PTH receptor occurred in association with PTH (1–34) and Gα_s, most likely as an active ternary complex (45). Furthermore, a distinct mechanism of sustained G protein activation in which the GPCR, β-arrestin, and G protein βγ-subunits form a complex has been reported in a study of the PTH receptor signaling (46). A comparable observation has been reported for the vasopressin 2 receptor, in which a similar complex seems to be involved in a sustained cAMP generation (47).

G protein-independent signaling comprises not only β-arrestin-mediated signaling processes (after G protein signaling) but also signaling events independent of G protein activation. One example is the well-studied angiotensin II type 1a receptor (AT₁R). Studies with cardiomyocytes demonstrated that strain, pressure, and shear stress induced AT₁R activation and subsequent downstream ERK activation (48). Another study with AT₁R mutants incapable to couple to heterotrimeric G proteins showed an angiotensin II-mediated activation of Scr and ERK (49). Furthermore, Zhai et al (50) studied the distinct physiological consequences of G protein-independent signal transduction via the AT₁R in the heart. They generated transgenic mice with cardiac-specific overexpression of a wild-type AT₁R or an AT₁R second intracellular loop mutant (AT₁-i2m). AT₁-i2m mice displayed a pronounced cardiac phenotype and found impaired AT₁-i2m-mediated hypertrophy in in vitro assays upon pharmacological inhibition of ERK and Src activity. However, it is unclear what underlying mechanism is mediating the G protein-independent activation of cytoplasmic effectors such as ERK and Scr. It remains to be elucidated what signaling (or other) molecule could potentially serve proximal functions similar to G proteins, but β-arrestin represents the most intriguing candidate (14, 50). Of note, β-arrestin-mediated signaling events can also take place in the absence of G protein activation. This has been demonstrated for the decoy receptor chemokine receptor-7 in which ligand binding activates MAPKs through β-arrestins in transiently transfected cells but does not result in activation of signaling pathways typical of G proteins (51).

Pharmacologically, GPCRs can be exploited in different ways, one of which is by targeting the orthosteric ligand binding site, which is the primary ligand binding site for the endogenous agonist (52, 53). The basic concept, that agonist binding induces a conformational change and switches the GPCR from an off to an on state, has been revised over the past several years because it is now appreciated that there is no equilibrium between on-off conformations. GPCRs rather adopt an ensemble of tertiary conformations that allow for new strategies that make use of other, spatially distinct allosteric binding sites to regulate diverse aspects of GPCR signaling (37, 54–56). Furthermore, the active state of a receptor is specified by the ligand-receptor complex and other small molecule or protein-protein interaction that allosterically constrains the conformations available to the receptor (57). Consequently, GPCRs can adopt multiple active conformations (58). Ligands that selectively bind and stabilize different GPCR conformations can modulate either extracellular (other ligands) or intracellular (signal transducing molecules) interactions and are called either orthosteric or allosteric modulators. They have evolved as important pharmacological tools to couple GPCRs to selectively activate distinct and/or overlapping signal transduction pathways. This phenomenon is termed biased signaling (52, 59–63). The phenomenon of signaling bias from the inside of the cell rather than from the ligand-binding side of the receptor is the newest concept (64). Key components for the coordination of signal transduction cascades are scaffolding proteins.

The importance of scaffolding proteins for GPCR signaling (Figure 1)

Figure 1. Schematic of GPCR/scaffolding protein interactions involved in GPCR signaling pathways (individual examples are discussed in the text).

Key proteins in the regulation of GPCR signaling are arrestins, PDZ proteins, and non-PDZ proteins, which all interact with the carboxyl terminal tail or the intracellular loops (mainly, but not exclusively, the third loop), respectively. Agonist activation of GPCRs leads to receptor phosphorylation which enables, yet is not always required, arrestin binding. A, Arrestin serves as adapter to link GPCRs to the endocytic machinery. Arrestin per se can also interact with multiple MAPKs and therefore is an important regulator of GPCR signaling. Furthermore, GPCR signaling from endosomes has been reported recently. B, The second group of proteins important for GPCR signaling is PDZ proteins. PDZ proteins have been shown to regulate various signaling pathways, such as the activation of ERK1/2, enhance or abolish the interaction with G proteins, hence modulating G protein-mediated signaling, facilitate the interaction with arrestin and PLC, and can activate Rho GTPase. The third group of scaffolding proteins that control GPCR signaling is the group of non-PDZ proteins. DAG, diacylglycerol. C, Non-PDZ proteins that interact with GPCRs have the ability to bind PDZ proteins but can also associate with PKA, Jak2 (to activate the Jak/STAT signaling pathway) and orchestrate Ras/Raf signaling cascades.

Scaffolding proteins are defined as proteins that bring together multiple proteins to assemble functional complexes thus modulating cellular signaling pathways. This is achieved by regulating numerous GPCR regulatory processes such as receptor stabilization, subcellular localization, and trafficking (sorting and recycling), which ultimately modulate downstream signaling events (65). Cytoplasmatic scaffolding proteins can be divided into three broad categories: arrestins, postsynaptic density 95/disc large/zona occludens-1 (PDZ) domain-containing proteins (PDZ proteins), and non-PDZ proteins (Table 1). GPCRs interact with such scaffolding proteins via their three intracellular loop domains and their carboxyl-terminus (20). As for the scaffolding proteins, the interaction with GPCRs is enabled, yet not exclusively, by multiple protein-protein interaction modules such as Scr homology 3, PDZ domains (66), or other critical determinants, ie, phosphorylation sites for β-arrestin binding.

Table 1.

Partial List of GPCR/Non-PDZ Scaffold Interaction Involved in GPCR Signaling

Scaffold Protein	GPCR	Effect	Reference
AKAP79	β2AR	Phosphorylation ↓	142–144
AKAP250	β2AR	Association with PKA ↑	143–145
	β1AR	Receptor recycling ↑	96, 146–147
AKAP5	GPR30	cAMP accumulation ↓	148
Jak2	AT1R	STAT phosphorylation ↑	149, 155–157
14-3-3	FSHR	cAMP accumulation ↓	159
	PAR4	Plasma membrane transport and signaling ↑	162
	α2AR	Ras/Raf signaling ↑	164, 165
	GPR15	Stability and cell surface expression ↑	163

Abbreviation: FSHR, FSH receptor.

Arrestin family

With a family of only four proteins in mammals and the lack of any recognizable sequence motif known to mediate protein-protein interactions, it is remarkable that β-arrestins bind to an astounding several hundred GPCRs to orchestrate cell signaling (67, 68). The arrestin family comprises the two visual arrestins (arrestin-1 and arrestin-4) that are confined to retinal cones and rods, the ubiquitously expressed nonvisual arrestins, β-arrestin-1 and -2 (arrestin-2 and -3) and was recently extended with a new class of arrestins called α-arrestins (69). α-Arrestins represent a fairly new member of the large family of scaffolding proteins known to be involved in the regulation of GPCR signaling. The members of the α-arrestin family comprise arrestin domain-containing1–5 and TXNIP/thioredoxin binding protein-2 (69). So far, α-arrestins have been shown to be involved only in the regulation of β₂AR ubiquitylation. Two possible models emerged from contradictory data; hence, the underlying mechanisms remains to be elucidated (69–72). On the other hand, the role of β-arrestins in the regulation of GPCR signaling has been well documented. The selectivity of β-arrestins for GRK phosphorylated GPCRs is considered to be regulated by two sensors for receptor activation and phosphorylation in arrestins that engage simultaneously (73–75). However, evidence suggests that even though many GPCRs must be phosphorylated to bind β-arrestins with high affinity, phosphorylation-independent β-arrestin binding is being reported for a growing number of GPCRs (76). Of note, it has been hypothesized that arrestins can form α-arrestin-β-arrestin heterodimers (69, 71). This may represent a potential new mechanism in, but also adds a new level of complexity to, the regulation of GPCR signaling.

There is now a perpetually growing list of nonreceptor β-arrestin-interaction partners that have been discovered over the past years, which revolutionized the role of β-arrestin within the cell (reviewed in references 77 and 78). Many of the identified β-arrestin-interacting proteins are themselves signaling molecules and include, but are not limited to the following: cAMP PDE4 (79), Yes (80), c-Jun N-terminal kinase 3, ERK1 and ERK2, cRaf1 (13, 81–84), phosphatidylinositol-3-kinase, Akt, and protein phosphatase 2A (77). The regulation of MAPK cascades by GPCRs involves a variety of mechanisms, including second messenger-dependent pathways, transactivation of classical receptor tyrosine kinases, and β-arrestin-scaffolded MAPK kinase complexes (78).

ERK1/2 signaling represents the best-studied G protein-independent and β-arrestin-dependent signaling pathway (81–83, 85). The relative stability of the receptor-β-arrestin complex regulates the duration of β-arrestin-dependent ERK activation and also influences the subcellular localization of ERK1/2 signals (86, 87). For example, in cell lines transfected with recombinant the PTH receptor (PTH1R) stimulates ERK1/2 via multiple mechanisms. PTH activation of the PTH1R results in ERK1/2 activation via a rapid and early G protein-dependent pathway meditated by PKA and PKC and a delayed and prolong G protein-independent pathway mediated by β-arrestins in human embryonic kidnay-293 cells (88). In nondifferentiated expression systems, the activation of β-arrestin-dependent and G protein-independent ERK1/2 signaling is also regulated by inverse agonists, indicating that multiple GPCR activation states contribute to G protein-dependent vs G protein-independent pathways (89). However, it is of note that differentiated cells exhibit a different mechanism of ERK1/2 activation independent of arrestin (90).

PDZ proteins

The human genome encodes for more than a hundred PDZ domain-containing proteins (PDZ proteins) (65). PDZ proteins consist of multiple PDZ domains that facilitate the binding to specific sequences within the intracellular domains of GPCRs, preferably but not exclusively, the C terminus. Such PDZ binding motifs comprise three to four amino acids and can be divided into three classes: class I (x-S/T-x-V/I/L), class II(-φ-x-φ), and class III(ψ-x-φ) (φ-hydrophobic amino acid, ψ-acidic amino acid) (66, 91). The GPCR PDZ binding motif allows for PDZ proteins to associate and scaffold multiprotein complexes to modulate GPCR function. PDZ proteins have been implicated in the regulation of virtually all receptor properties (16–17, 92). However, this review focuses on their implications in GPCR signaling/desensitization (Table 2). Prototypical PDZ proteins that have been shown to interact with various GPCRs include the following: the multiple PDZ domain protein (MUPP-1), members of the membrane-associated guanylate kinase family, such as synapse-associated protein 97 (SAP97), postsynaptic-density protein 95 (PSD-95), synapse-associated protein 102, WW and PDZ domain-containing proteins (MAGIs), sorting nexin 27 (SNX27), and members of the Na⁺/H⁺ exchanger regulatory factor (NHERF) family like NHERF1–4.

Table 2.

Nonexhaustive List of GPCR/PDZ Scaffold Interaction Involved in GPCR Signaling

GPCR	PDZ Protein	Effect	Reference
mGluR1a	PIST (CAL; GOPC)	ERK activation ↓	210
mGluR5	NHERF2	Ca2+ signaling ↑	211
β1AR	GIPC	ERK activation ↓	212
	MAGI-3	ERK activation ↓	97
	SAP-97	PKA-mediated receptor phosphorylation ↑	96
	PIST	ERK activation ↓	104
β2AR	SNX27	Receptor recycling ↑	98
	MAGI-3	ERK activation ↓	213
	NHERF1	Plasma membrane recycling ↑	93, 94
α2AR	SNX27	Functional resensitization ↑	106, 107
	Spinophilin	Cell surface expression ↑, phosphorylation ↓, MAPK signaling ↓, Ca2+ signaling	132–134, 136, 137
LPA2	MAGI-3	ERK activation ↑, ρ activation ↑	214
	NHERF2	Interaction with PLCβ3 ↑, IP3 signaling ↑, ERK activation ↑	119, 120
MT1R	MUPP-1	Coupling with Gi protein ↑, inhibition of adenylyl cyclase activity ↑	109
PTH1R	NHERF1 (EBP50)	Coupling to and activation of Gq protein ↑, interaction with β-arrestin-2 ↓, desensitization ↓, cAMP signaling ↑	122, 215, 216
	NHERF2	Interaction with and activation of PLCβ ↑, adenylyl cyclase activity through stimulation of Gi/o proteins	121, 215
D1R	PSD-95	cAMP signaling ↓	217
D2R	GIPC	G protein coupling↓	115, 116
	Spinophilin	Signaling ↑	133
P2Y1R	NHERF2	Interaction with PLCβ ↑, agonist-induced Ca2+ signaling ↑	124
5-HT2AR	PDS-95	IP3 signaling ↑	128
	PDZK1/NHERF3	IP3 signaling ↑	127
	SAP97	IP3 signaling ↑, ERK activation ↑	207
GPR30	SAP97/PSD-95	cAMP signaling ↓	148
SSTR5	PDZK1/NHERF3	Interaction with PLCβ3 ↑	118
	PIST (CAL, GOPC)	Trafficking, cell surface expression ↓	102
CRFR1	SNX27	Receptor recycling ↑	102
	PDZK1/NHERF3	ERK activation ↑	127
	SAP97	ERK activation ↑	126
LHR	GIPC	ERK activation ↑	114
GHRHR	PICK1	Cell surface expression ↑, cAMP signaling ↓	218
M3R	Spinophilin	Signaling ↓	139
μ-OR	Spinophilin	Endocytosis ↑	140
IP	PDZK1/NHERF4	Cell surface expression ↑, cAMP signaling ↑	129

Abbreviations: D1R, D1 dopamine receptor; EBP50, ezrin-radixin-moesin-binding phosphoprotein-50; GHRHR, GHRH receptor; IP, I prostanoid receptor; LPA2, lysophosphatidic acid receptor 2; mGluR1/5, metabotropic glutamate receptor 1/5; P2Y1R, purinergic receptor type 1.

The β₁AR is by far the best-studied receptor when it comes to PDZ protein interactions and their implication in the regulation of receptor activity. β₁AR is known to bind to various PDZ proteins such as cystic fibrosis transmembrane conductance regulator-associated ligand (CAL), PDZ protein interacting specifically with TC10 (PIST)/golgi-associated PDZ and coiled-coil domains-containing protein (GOPC), PSD-95, MAGI-2, GAIP-interacting protein C terminus (GIPC), MAGI-3, SAP97, and SNX27 (93–100). Although the interaction with MAGI-3 terminates β₁AR-mediated ERK1/2 activation with no effect on agonist-stimulated cAMP formation or agonist-mediated internalization, the strong binding of MAGI-2 promotes β₁AR internalization (100). The association of the Golgi-associated PDZ protein CAL to the β₁AR results in intracellular receptor retention (101). PSD-95, another PDZ protein, can compete with CAL for the binding to the β₁AR, which allows for β₁AR transport to the cell surface but also antagonizes β₁AR endocytosis (95, 101).

A similar phenomenon is reported for the somatostatin receptor (SSTR)-5, but in this case, NHERF released SSTR5 from trans-Golgi network (TGN) retention (102, 103). More recent studies analyzed these interactions in more detail and determined that CAL/PIST retains β₁AR in the TGN and that CAL/PIST overexpression reduces the MAPK pathway by β₁AR agonists. Upon agonist stimulation CAL/PIST relocates from the TGN to an endosomal compartment, thus contributing to β₁AR sorting through anterograde and retrograde trafficking (104). A third PDZ protein involved in SSTR5 trafficking, particularly recycling, is SNX27. The functional resensitization of SSTR5 is shown to be regulated by SNX27 (102). Consequently, the interaction of SSTR5 with PIST/PDZ kinase PDZ domain-containing 1 (PDZK1)/SNX27 may contribute to an on-demand cell surface delivery of the receptor (105). This process might strongly influence the ability of somatostatin to regulate insulin secretion from the pancreas or GH secretion from the pituitary. SNX27 also interacts with both the β₁AR and β₂AR (98, 106, 107). SNX27 mainly functions to receptor recycling back to the plasma membrane, thereby increasing the availability of functional receptors for agonist binding on the cell surface. Thus, recycling and resensitization represent additional critical steps in the regulation of GPCR signaling.

Another SSTR subtype, SSTR3, was shown to interact with MUPP-1, which enables the incorporation of SSTR3 at tight junctions in a PDZ-dependent manner, and allows SSTR3 to control transepithelial conductance (108). Furthermore, MUPP-1 binds to the melatonin 1 receptor (MT1R) and abolishes Gα_i-mediated signaling by interfering with MT1R-Gα_i protein interaction (109). Another binding partner of MUPP-1 is the serotonin 2C receptor. This interaction facilitates receptor phosphorylation and resensitization in a PDZ motif-dependent manner (110–112).

Signal specificity and diversity for distinct receptor families, such as GPCRs, can potentially be regulated by membrane trafficking. Receptors targeted to either lysosomal or recycling pathways generate different signaling profiles that result in either the permanent termination of G protein signal or resensitization of G protein signals. Thus is due to postendocytic sorting of receptors back to the cell surface, a process that is regulated by PDZ proteins (113). In this respect, GIPC represents a PDZ protein that plays an important role in regulating LH receptor (LHR) trafficking through the endocytic compartment as a consequence of its interaction with the C-terminal tail of the receptor. Specifically, GIPC routes the LHR to pre-early endosomes for recycling rather than to early endosomes (114). This rerouting of the LHR from early endosomes to pre-early endosomes or vice versa appears to spatially reprogram MAPK signaling responses (114). In addition, dopamine receptor (D2R) and dopamine receptor 3 signaling is regulated by GIPC. For example, in neuroendocrine cells the D2R recruits GAIP [regulator of G protein signaling (RGS)-19], which is required for GIPC binding. Thus, GIPC might serve as a molecular adaptor between GPCRs and RGS proteins to facilitate the attenuation of D2R signaling (115, 116).

NHERF proteins have been linked to the regulation of endocrine GPCR signaling and trafficking processes. One example comprises the formation of a ternary complex of the PTH1R, NHERF1, and β-arrestin-2 (117). This study suggests a model by which NHERF1 serves as an adaptor protein to recruit β-arrestin-2 into close proximity to the PTH1R, thus enabling β-arrestin-2 recruitment upon PTH1R activation. NHERFs also appear to be critical for the interaction with PLC-β. For example, NHERF3 (PDZK1) enables SSTR5 coupling to PLC-β3 (118), NHERF2 connects lipophosphatidic acid receptor 2 to the activation of PLC-β3 (119, 120), and NHERF1 interactions with PTH1R link the receptor to PLC-β (121). This interaction causes a shift in the second-messenger response from adenylyl cyclase activation to the hydrolysis of phospholipids (121) and inhibits PTH1R desensitization (122).

For the thromboxane A2 receptor, agonist stimulation promotes the interaction of NHERF1 (ezrin-radixin-moesin-binding phosphoprotein-50) with Gα_q (123) and antagonizes thromboxane A2 receptor-mediated activation of Gα_q and PLC-β1. NHERFs not only connect receptors to PLC-β but also regulate calcium signaling by the purinergic receptor type 1. This is accomplished by the interaction of NHERF2 (124). NHERF1 is also required for efficient plasma membrane recycling of the β₂AR (93, 125). Furthermore, GPCR/PDZ protein interactions play a crucial role in the regulation of inositol phosphate signaling (126–128). This has been extensively studied for the 5-hydroxytryptamine receptor (5-HT_2AR) with reports on multiple PDZ proteins, including NHERF3/PDZK1, SAP97, and PSD-95, being involved in regulating Gα_q signaling. The interplay between these three PDZ proteins and the 5-HT_2AR is still unclear and remains to be elucidated. NHERF4/PDZK2 also interacts with the human I prostanoid receptor and agonist activation of the receptor leads to PKA- and PKC-mediated phosphorylation of NHERF4/PDZK2 (129).

Spinophilin is another PDZ domain-containing protein that has been shown to interact with more than 30 diverse partner proteins, including GPCRs (130). However, most the GPCR-mediated interactions with spinophilin are mediated by a spinophilin receptor-interacting domain, which is described to associate with an interaction site within the third cytoplasmic loop of the α₂-adrenergic receptor (α₂AR) and D2R (131–133). This interaction contributes to the cell surface stabilization of adrenergic receptors (134). Agonist occupancy of α_2A/2BAR enhances spinophilin but also antagonizes GRK2 and β-arrestin binding to the receptors (135). Spinophilin blocks receptor phosphorylation by recognizing the complex of α₂AR with Gβγ and thereby competes with GRK2 for the binding to this complex. Consequently, spinophilin reduces receptor endocytosis and the stimulation of MAPKs (136). It has also been shown that spinophilin can actively regulate α_1BAR function because it recruits RGS2 to the α_1BAR/G_q signaling complex to form a heterotetrameric complex that is implicated in the regulation of Ca²⁺ signaling by α-agonists in cultured cells (137). In vivo studies with the α_2AAR in spinophilin-null mice reveal the impact of spinophilin on α₂AR-evoked cardiovascular responses (138). In mice lacking spinophilin, α₂AR stimulation leads to an enhanced and prolonged hypotensive, bradycardic, and sedative-hypnotic responses. Eliminating spinophilin expression in native tissue also allowed for enhanced receptor/G protein coupling efficiency, which contributes to the sensitization of α₂AR-mediated responses in vivo. Furthermore, spinophilin is a negative regulator of the M3-muscarinic acetylcholine receptor (m3 AchR)-mediated signaling because it is able to recruit RGS4 in an agonist-dependent manner to the m3R signaling complex (139) and promotes μ-opioid receptor endocytosis (140).

Non-PDZ proteins

A-kinase anchor proteins (AKAPs)

AKAPs were first identified as scaffolding proteins in 1999 (141). Because AKAPs coordinated the subcellular localization of second messenger-regulated enzymes such as PKA, PKC, and protein phosphatases, they orchestrate the regulation of these signaling pathways. AKAP79 and AKAP250 were identified as interaction partners for the β₂AR (142–145). Specifically, AKAP79 promoted β₂AR phosphorylation and downstream mitogenic signaling, whereas AKAP250 regulated β₂AR desensitization via its association with PKA. AKAP79 was also found to interact with the β₁AR/SAP97 protein complex (96), and SAP97 was previously shown to play a critical role in β₁AR recycling (93, 94). This effect on β₁AR desensitization was found to be dependent on the formation of a SAP97-AKAP79-PKA complex that was tethered by SAP97 to the type I PDZ motif of the β₁AR (96, 146, 147). Recently another GPCR, GPR30, was added to the list of AKAP-interacting GPCRs. GPR30 was shown to form a PDZ motif-dependent plasma membrane complex with a membrane-associated guanylate kinase (likely SAP97 or PSD-95) and AKAP5, which led to the constitutive attenuation of cAMP production and retained the receptor at the plasma membrane (148).

Janus kinase (Jak)-2

The Jak signal transducers and activators of the transcription (STAT) signaling pathway is typically activated by cytokine or growth factor receptors but not by GPCRs. However, stimulation of the AT₁R has been found to activate not only G protein-dependent pathways but also the Jak/STAT signaling pathway (149–155). The ability of the AT₁R to regulate Jak/STAT signaling is dependent on a direct interaction of AT₁R with the tyrosine kinase Jak2 (149, 155–157). The scaffolding of Jak2 on the AT₁R was dependent on a distinct C-terminal tyrosine motif (YIPP) in the AT₁R C terminus (156) that is required for Jak2 phosphorylation of the transcription factor STAT and the recruitment of STAT1 into a complex with the AT₁R (149, 155).

14–3-3 proteins

Isoforms of 14–3-3 proteins have been shown to play key roles in signal transduction pathways. 14–3-3 proteins exhibit the characteristics of scaffolding proteins (158) and represent a potential mechanism by which GPCR-mediated signaling is modulated. Examples include the association of 14–3-3 with the FSH receptor in a follitropin-dependent manner that results in a modest decrease of follitropin accumulation (159). Furthermore, 14–3-3 proteins recognize and mask retinoid X receptor motifs to direct endoplasmic reticulum (ER)/Golgi export of multimeric proteins (160, 161). The protease-activated receptor (PAR)-4 encodes an ER retention motif and is usually retained in the ER (162). However, PAR4 evades ER retention by forming a heterodimer with PAR2, which allows interaction with 14–3-3ζ. Hence, the receptor is transported to the plasma membrane and exhibits substantially enhanced signaling properties. Similarly, phosphorylation-dependent 14–3-3 binding to G protein-coupled receptor 15 (GPR15) has been shown to increase GPR15 stability and increased cell surface expression due to the suppression of the dibasic (retinoid X receptor) ER retention signal (163). The 14–3-3ζ isoform also interacts with the α₂AR and couples the α₂AR to the activation of the Ras/Raf cascade via the G protein βγ-subunit (164, 165). Moreover, phosphorylated Raf functions to antagonize receptor-14–3-3ζ interactions, indicating that the 14–3-3 proteins interact with inactive α₂ARs to facilitate the activation of the Ras/Raf cascade after the agonist binding. After the agonist activation, the βγ-subunit and the 14–3-3 protein subsequently dissociate to enable Ras/Raf signaling (164, 165).

GPCRs in the endocrine system: implications in pathological states

Major endocrine systems comprise the hypothalamic-pituitary-thyroid axis, the hypothalamic-pituitary-gonadal axis, the hypothalamic-pituitary-adrenal (HPA) axis, the renin-angiotensin system, and energy homeostasis. Key players in these systems are the GPCRs that, upon hormone binding, mediate the downstream response in the target organ as well the GIPs that modulate and/or control downstream signaling profiles. Either inappropriate hormone release or response to signaling, usually caused by mutations, results in endocrine disorders (reviewed in references 3 and 166). A plethora of GPCRs and their associated interacting proteins have been implicated in endocrine diseases. Thus, it is impossible to address all of them within the scope of this review. Herein we will focus on a few selected examples, such as osteoporosis, cancer, diabetes, obesity stress, and anxiety because these diseases represent prevalent human health risks with hundreds of millions of people affected worldwide.

Obesity and diabetes

Obesity represents a critical predisposition to insulin resistance and the subsequent development of type 2 diabetes mellitus (T2DM). Targets for therapeutic interventions include several members of the GPCR superfamily and their ligands as well as various GPCR-scaffolding proteins (167). Ghrelin, with its critical role in energy homeostasis, and its receptor, the GH secretagogue receptor [ghrelin receptor (GHSR1)] are of great significance for the treatment of obesity (168). GHSR1 signals through Gα_q/11, Gα_i/o, and Gα_12/13 as well as β-arrestin-based scaffolds. GHSR1 enhances appetite, GH release, and modulation of reward and cognitive functions. Mutagenesis studies revealed different receptor conformations that support signaling bias at the GHSR1 and could prove beneficial for the development of future antiobesity drugs (168, 169).

The metabolic syndrome is a condition that can lead to diabetes. Normal blood glucose levels in a healthy individual depend on the regulation of insulin secretion from pancreatic β-cells and a malfunction in insulin secretion causes T2DM (170). One GPCR involved in the maintenance of proper blood glucose levels is, among many others (171), the m3 AChR. Novel strategies in the treatment of T2DM include the enhancement of β-cell m3 AChR signaling, which involves G protein-dependent and G protein-independent such as β-arrestin-dependent signaling pathways (172). Studies with transgenic mice that expressed an m3 AChR-based (designer receptor exclusively activated by designer drugs, a receptor unable to bind the endogenous muscarinic receptor ligand ACh but can be efficiently activated by clozapine-N-oxide, an otherwise pharmacologically inert compound) (173–175). Designer receptors exclusively activated by designer drugs provide great potential in understanding the contribution of GPCR signaling, particularly biased signaling and scaffolding proteins as critical determinants of disease pathology. Because it has been discovered that m3 receptors can potentiate insulin release through a β-arrestin-dependent mechanism, targeting muscarinic receptor signaling with biased agonists selective for the β-arrestin-dependent pathways could prove therapeutically beneficial. Importantly, reduced side effects due to minimized G protein-dependent calcium signaling would be expected. Consequently, biased ligands that direct signaling of the m3 AChR via β-arrestin signaling would be expected to potentiate the insulin release at β-islets (176).

In addition to β-arrestin, spinophilin represents another scaffolding protein capable of regulating m3 AChR signaling (139, 177). It has been identified as a negative regulator due to its selective inhibition of m3 AChR-mediated signaling and insulin release, which is mediated by RGS4 (178). Another GPCR expressed in pancreatic β-cells is the SSTR5. The availability of SSTR5 at the plasma membrane in these cells has been shown to be regulated by the scaffolding protein PIST (179). Accordingly, PIST plays a critical role in the SSTR5-mediated regulation of insulin secretion. In summary, scaffolding proteins, may serve as potential targets for therapeutic intervention aimed to modulate β-cell m3 AChR or SSTR5 function for the treatment of T2DM (139, 177–179).

Osteoporosis

Reduced bone mineral density in elderly people leads to increased susceptibility to fracture, an age-related metabolic disorder, called osteoporosis (180). One of the two GPCRs involved in calcium homeostasis is the PTH1R, mainly expressed in kidney and bone. PTH signaling exerts multifaceted effects on bone (181, 182). It simultaneously promotes new bone formation by osteoblasts and bone loss through osteoclastic bone resorption (183). The effect of PTH on bone is complex and critically regulated by the kinetics of PTH1R activation because anabolic and catabolic effects of PTH are coupled. Hence, the PTH1R demonstrates the immense diversity of the signaling mechanisms that can originate from a single GPCR (182, 184). Specifically, the PTH1R represents an important target for the treatment of osteoporosis by using novel therapeutic strategies that selectively bias the activation of one specific signaling pathway over another. Thus, targeting exclusively the β-arrestin pathway will allow for selective targeting of anabolic bone formation yet at the same time exclude the catabolic/calcitrophic effects of PTH1R activation, such as bone resorption (184, 185). In conclusion, the PTH-β-arrestin signaling pathway may represent a novel pharmaceutical opportunity in the battle against osteoporosis.

Another regulatory component of PTH1R signaling is its association with the PDZ proteins NHERF1 and NHERF2 (121). NHERF1/2 association influences the specificity of G protein and/or β-arrestin coupling. Although NHERF1 binding to PTH1R allows for G_q and PLC activation without affecting its ability to stimulate G_s and adenylyl cyclase, NHERF2 enables coupling to G_i and G_q but prevents coupling to G_s. Consequently, NHERF2 enables the PTH1R to switch from activating AC to PLC (121). In summary, NHERF-mediated signaling provides a mechanism by which cells regulate responses to PTH in a tissue-specific manner but may also be a mechanism by which the cells specify sites of second messenger generation.

Endocrine tumors

GPCRs and their signaling networks have recently emerged as crucial players in aberrant tumor growth and metastasis. In fact, most gain-of-function mutations in GPCRs and the Gα subunit of G proteins were identified in endocrine tumors, which cause uncontrolled hormone secretion (reviewed in reference 186). One prototypical example is the TSHR, found in thyroid carcinomas and almost all thyroid adenomas. A characteristic of these active TSHR mutants is a constant Gα_s-dependent activation of adenylyl cyclase and Gβγ-dependent stimulation of phosphatidylinositol-3-kinase and MAPKs in thyrocytes (187). Because GPCRs regulate second messengers and their downstream targets, they can control cell survival, growth, and migration by activating MAPK cascades. It is appreciated that such MAPKs are critical for cell proliferation and metastasis. Consequently, their dysregulation is closely linked to human malignancies (188). In addition to the activation of second-messenger signals, GPCR-scaffolded proteins such as β-arrestin and PDZ proteins may represent potential targets for cancer treatment. β-Arrestins are involved in several cancer-related signal networks via multiple different receptor subtypes such as the endogenous prostanoid (EP)-1–4 receptors, endothelin type A receptor, lysophosphatidic acid (LPA) receptor, and β₂AR (189–192). The EP4 receptors have been demonstrated to mediate prostaglandin-induced A549 lung cancer cell migration (190). Prostaglandin E₂ (PGE₂) stimulation enhances c-Scr activation, which suggests that blockage of c-Scr activity may be a target for the inhibition of PGE₂-mediated lung cancer cell migration. Direct EP4 effectors of migration include β-arrestins and heterotrimeric G_s proteins. Notably, PGE₂-mediated c-Scr activation and A549 cell migration is abolished when β-arrestin was knocked down. Consequently, the EP4-β-arrestin1 signalosome represents a promising drug target for the treatment of lung cancer. Specifically, selective EP4 antagonists may prove beneficial (190). Furthermore, EP receptors are also involved in the transactivation of epithelial growth factor receptor, which is mediated by PGE₂-stimulated EP2 and EP4 receptors. EP receptors are implicated in mouse skin papilloma development (189) and colorectal carcinoma cell migration and metastasis (193), which not only involves both β-arrestin1 and c-Src. Likewise, β-arrestins mediate endothelin type A receptor-stimulated epithelial growth factor receptor transactivation via c-Src, as well as effects on β-catenin, in ovarian cancer cells (192).

LPA receptors represent yet another example for β-arrestin-mediated signaling in cancer cells (191). In a recent study, Li et al (191) have determined a novel role for β-arrestin/Ral signaling in mediating LPA-induced breast cancer cell metastasis. Prostate cancer cell progression has been shown to be induced by β-arrestin2-mediated β₂AR signaling (194). Herein β₂AR activation promoted increasing levels of cAMP and ERK1/2 activation, resulting in enhanced cell proliferation and migration in prostate cancer cells. Furthermore, β₂AR activation augmented the formation of a β-arrestin2/c-Scr complex, which enables the use of a c-Scr inhibitor as a potential therapeutic tool to block the aforementioned complex formation as well as subsequent cell proliferation (194). However, β-arrestins not only activate proliferative signaling pathways but also engage in various pathways that suppress tumor growth and metastasis. This highlights their important therapeutic potential for cancer therapy (195). The androgen-dependent prostate cancer cells represent an example in which β-arrestin2 recruits the E3 ligase mouse double minute-2, which subsequently leads to receptor ubiquitylation and down-regulation (196, 197).

PDZ proteins have also been closely connected to human malignancies (198, 199). Spinophilin expression has been shown to determine cellular growth in colorectal cancer (198). However, little, if anything, is known about GPCRs and their interactions with spinophilin in colon cancer growth. Nevertheless, because various GPCRs such as EP2, EP4, LPA₁, endothelin, and PAR1 receptors have been implicated in colon cancer (186) and spinophilin is known to interact with them, it is likely that spinophilin interactions may be important for the antagonism of cancer cell growth (132–134, 136, 137, 139). Another PDZ protein that is linked to cancer is NHERF1 (200, 201). Mutations in NHERF1 have also been associated with breast cancer (200), and NHERF1 is down-regulated during the development of colon cancer (201). In pancreatic cancers, the PDZ protein GIPC1/synectin is often overexpressed (202). Importantly, blocking synectin's PDZ-binding potential in a mouse model of pancreatic tumor invasion results in the inhibition of tumor growth (203). Although the roles of the PDZ interactions of NHERF1, GIPC, and their cognate GPCRs have yet to be determined, it is easy to envisage that such GPCR/PDZ protein interactions are likely of importance in cancer growth and development. Our rapidly growing knowledge of the expression status and functional activity of GPCRs along with their associated interacting proteins in various cancer types and their exceptional drugability make them ideal candidates for cancer prevention and treatment in the near future (186).

Stress and anxiety

Corticotropin-releasing factor (CRF) is a neuropeptide that regulates the physiological response to stress by initiating the HPA axis stress response as well as regulating prefrontal cortical stress responses. Consequently, CRF is also implicated in the manifestation of depression (204). Its involvement in the regulation of stress and anxiety presents a functional link between endocrine signaling and neurobehavioral responses (205). Of note, serotonergic dysfunction is also well established in mood disorders because the serotonergic system is regulated by the HPA axis. CRF and 5-hydroxytryptamine exert their biological effects by activating a specific group of GPCRs, the CRF receptors (CRFRs) and 5-hydroxytryptamine receptors. These receptors represent promising targets for the treatment of mood disorders (206).

Recently it has been reported that the preactivation of CRFR1 sensitizes of 5-HT_2AR inositol phosphate signaling by a mechanism that requires intact receptor endocytosis, recycling, and increased 5-HT_2AR expression at the cell surface (207). Both CRFR1 and 5-HT_2AR encode a PDZ binding motif at their carboxyl-termini, and deleting of either PDZ motif resulted in a loss of heterologous sensititization of 5-HT_2AR signaling that is associated with the inability of the 5-HT_2AR to be mobilized to the cell surface. Even though multiple PDZ proteins (PDZK1, PSD-95, SAP97) have been identified as CRFR1 and/or 5-HT_2AR interacting partners, the precise PDZ protein involved in the CRFR1/5-HT_2AR cross talk remains to be determined (127, 128, 208). Nevertheless, the interaction of PDZ proteins with GPCRs represents a novel regulatory mechanism for functional receptor cross talk and highlights the importance of PDZ proteins in the regulation of GPCR function. Importantly, the down-regulation of 5-HT_2AR-mediated signaling is a critical process achieved by second generation atypical antipsychotics (209). Thus, the selective prevention of a PDZ protein/5-HT_2AR interaction may result in a down-regulation of 5-HT_2AR surface expression and subsequently decreased 5-HT_2AR signaling. This may provide a novel therapeutic approach for the treatment of mood disorders.

Conclusion

The role of agonist stimulation for the association of a GPCR with a scaffold protein still remains poorly understood because it likely plays a critical role for the regulation of receptor signaling. Agonist stimulation causes a major change in GPCR conformation which leads to a significantly enhanced association of G proteins with the receptor's intracellular domains. However, although some receptor/scaffold interactions have been found to be enhanced by agonist stimulation (β₂AR/NHERF, AT1R/Jak), others do not require agonist binding (β₂AR/AKAP). The underlying mechanism contributing to these differences exist, and the physiological significance of these interactions has yet to be fully explored and will require the transgenic overexpression and knockout of these proteins. Because scaffolding proteins have the ability to enhance protein-protein interactions yet do not possess an intrinsic activity of their own, studying their activity is nearly impossible in a nonphysiological setting. Furthermore, it still remains very challenging to fully understand cellular functions of scaffolding proteins. This is due to their ability to bind multiple proteins and generate very complex signal transduction complexes. In general, studying the interaction of more than two proteins can be extremely complicated. However, future research aimed at understanding how various binding partner bind simultaneously to a scaffold is essential. If these proteins do not bind simultaneously, it will be particularly interesting to understand both the order and kinetics of binding and whether one binding partner modulates the association/dissociation of other proteins.

Over the past few decades, it has become increasingly evident that GPCR activity is critically regulated by a plethora of interacting proteins, including proteins that are required for the scaffolding of macromolecular complexes. The formation of these complexes is not only essential for regulating the normal cell biology of GPCRs but are also important for the modulation of the normal and pathophysiological responses that contribute to the regulation of human health. A growing number of human diseases have been found to result from a dysfunction of such GPCR-scaffolding protein interactions. Therefore, delineating the mechanism of GPCR dysfunction represents an important strategy to either therapeutically restore or inhibit receptor activity. The current decade of GPCR research has revolutionized our understanding of GPCR signaling and provides new avenues for therapeutic molecule intervention in endocrine disease processes.

Funding Statement

This work was supported by Canadian Institutes of Health Research Grant MOP-62738 (to S.S.G.F.), and S.S.G.F. is a Career Investigator of the Heart and Stroke Foundation of Ontario, Canada. C.W. is the recipient of a Bell Mental Health Research Training Award from the Brain Canada Foundation and Bell Canada.