In Vivo Metabolic Roles of G Proteins of the G_i Family Studied With Novel Mouse Models

Abstract

G protein–coupled receptors (GPCRs) are the target of ~30% to 35% of all US Food and Drug Administration–approved drugs. The individual members of the GPCR superfamily couple to 1 or more functional classes of heterotrimeric G proteins. The physiological outcome of activating a particular GPCR in vivo depends on the pattern of receptor distribution and the type of G proteins activated by the receptor. Based on the structural and functional properties of their α-subunits, heterotrimeric G proteins are subclassified into 4 major families: G_s, G_i/o, G_q/11, and G_12/13. Recent studies with genetically engineered mice have yielded important novel insights into the metabolic roles of G_i/o-type G proteins. For example, recent data indicate that G_i signaling in pancreatic α-cells plays a key role in regulating glucagon release and whole body glucose homeostasis. Receptor-mediated activation of hepatic G_i signaling stimulates hepatic glucose production, suggesting that inhibition of hepatic G_i signaling could prove clinically useful to reduce pathologically elevated blood glucose levels. Activation of adipocyte G_i signaling reduces plasma free fatty acid levels, thus leading to improved insulin sensitivity in obese, glucose-intolerant mice. These new data suggest that G_i-coupled receptors that are enriched in metabolically important cell types represent potential targets for the development of novel drugs useful for the treatment of type 2 diabetes and related metabolic disorders.

G protein–coupled receptors (GPCRs) modulate the activity of many key metabolic functions that regulate glucose and energy homeostasis (1, 2). Metabolically relevant tissues and cell types express dozens of different GPCRs endowed with distinct G protein coupling properties (see below) (3). Because of the diversity of G protein subunits and downstream effector molecules, the pattern of cellular changes caused by activation of a specific GPCR in a particular cell type is quite complex. The precise nature of these responses critically depends on which type of G proteins are activated and which specific effector molecules are expressed in a specific target tissue or cell type. In addition, the magnitude of the observed cellular changes is greatly affected by the relative concentrations of the various components of the different receptor/G protein signaling pathways.

At the molecular level, the binding of agonist ligands to GPCRs leads to the selective activation of 1 or more distinct classes of heterotrimeric G proteins which consist of an α-subunit (Gα) that is tightly bound to a βγ-complex. In the inactive form of the G protein heterotrimer, a GDP molecule is bound to the α-subunit. Ligand-activated GPCRs trigger confirmational changes in G protein structure that promote the exchange of guanosine diphosphate (GDP) for guanosine triphosphate (GTP) on Gα. GTP binding leads to the dissociation of Gα from the receptor as well as from the βγ-complex. The GTP-bound α-subunits are then able to modulate the activity of distinct downstream effector enzymes and/or ion channels. Free βγ-complexes can also contribute to G protein modulation of cellular functions (see below). To date, more than 20 different mammalian Gα-subunits (encoded by 17 distinct genes) have been identified. On the basis of amino acid and functional similarity of the Gα-subunits, G proteins are generally grouped into 4 major families: G_s, G_i/o, G_q/11, and G_12/13 (4).

Heterotrimeric G Proteins of the G_i/o Family

The G_i/o family of G proteins represents the largest and most diverse G protein family (4, 5). The α-subunits contained within G_i/o heterotrimeric G proteins include Gα _i1, Gα _i2, Gα _i3, Gα _o, Gα _t1, Gα _t2, Gα _gust, and Gα _z. G_i1, G_i2, and G_i3 are widely expressed by most cell types. Gα _o, which exists in 2 splice variants (Gα _o1 and Gα _o2), is predominantly found in neurons. Gα _t1 and Gα _t2 (t stands for transducin) are expressed in rod and cone cells of the eye, respectively. Gα _gust (gust stands for gustducin) is primarily present in the gustatory epithelium. Gα _z is expressed in neurons and a limited number of peripheral cell types (4).

The best-known function of activated Gα _i/o-subunits is their ability to directly inhibit the activity of different isoforms of adenylyl cyclase. Rhodopsin/opsin-mediated activation of the transducin α-subunits (Gα _t1/_t2) leads to the stimulation of cyclic guanosine monophosphate-phosphodiesterase (4).

Like Gα-subunits, Gβγ-complexes can also regulate the activity of various effector molecules (6, 7). At present, 5 different G protein β- and 12 distinct γ-subunits have been identified in the human and mouse genomes (6). Gβγ-complexes are extremely stable and, therefore, function as 1 functional unit.

Since the expression levels of Gα _i/o proteins are relatively high, receptor-mediated activation of G_i/o is predicted to trigger the release of relatively large amounts of free βγ-complexes. For this reason, many of the cellular responses observed after activation of G_i/o involve βγ-mediated functions (4, 6). However, the ability of G_i/o to stimulate βγ-dependent signaling is also affected by other factors unrelated to the high expression levels of the Gα _i/o-subunits (6). Effector proteins whose activity is regulated by free βγ-complexes include, for example, inwardly rectifying K⁺ channels (activation), different isoforms of phospholipase Cβ (activation) and adenylyl cyclase (activation or inhibition), N- and P-type Ca²⁺ channels (inhibition), and phosphoinositide 3 kinase γ (activation) (4, 6).

Except for G_z, all members of the G_i/o family contain a Cys residue close to the C-terminus of their α-subunits (4). In the presence of pertussis toxin (PTX), which is produced and secreted by Bordetella pertussis, an ADP-ribose moiety becomes covalently attached to this Cys residue (8). As a result of this structural change, GPCRs are no longer able to activate PTX-modified G_i/o proteins (4). For this reason, PTX has served as a very valuable tool for studying the in vitro functional roles of G_i/o signaling. For simplicity, I will use the term “G_i,” rather than “G_i/o,” throughout this review.

Novel Mouse Models to Study the In Vivo Metabolic Roles of G_i Signaling

This review will primarily focus on the in vivo metabolic roles of G_i activity in several metabolically relevant cell types. It is well known that G_s- and G_q/11-mediated signaling pathways also have major effects on whole body glucose and energy homeostasis (for a recent review, see (1)). The data summarized in this review mostly deal with the use of 2 mutant mouse models that have led to many novel insights into the in vivo metabolic roles of G_i-type G proteins. More than a decade ago, Regard et al. (9) reported the development of a mouse line that allows for the conditional inactivation of G_i proteins in a cell-type–specific fashion. In this mouse strain, a cDNA encoding the S1 catalytic subunit of PTX (S1-PTX) was inserted into the ROSA26 locus such that S1-PTX could be expressed upon excision of a floxed transcriptional/translational stop sequence by Cre recombinase. In the following, I refer to these mice simply as Rosa-LSL-PTX mice (LSL=loxP-stop-loxP). Rosa-LSL-PTX mice were first used to selectively inactivate G_i-type G proteins in pancreatic β-cells (9). More recently, this approach has been extended to several other metabolically important cells types including mouse hepatocytes (10) and adipocytes (11). Importantly, the use of Rosa-LSL-PTX mice makes it possible to inactivate all members of the G_i family, except for G_z, in a particular cell type. Attempts to achieve this goal by employing classical gene knockout strategies would be extremely costly and time-consuming.

While Rosa-LSL-PTX mice are being used to disrupt G_i signaling in a conditional fashion, a recently developed mouse strain makes it possible to selectively stimulate G_i in a cell- or tissue-specific manner in vivo. Specifically, Zhu et al. (12) generated a novel mouse strain in which the coding sequence of a G_i-coupled designer GPCR (hM4Di) (13) was inserted into the ROSA26 locus. In this mouse line, the hM4Di coding region is preceded by an LSL sequence. As a result, the hM4Di designer receptor is only expressed in cells that produce Cre recombinase.

The hM4Di receptor is a member of a new class of designer GPCRs known as DREADD (Designer Receptors Exclusively Activated by a Designer Drug) (1, 13, 14). The most commonly used DREADDs are mutant muscarinic receptors that show little or no activity in the presence of acetylcholine, the endogenous muscarinic receptor agonist (1, 13, 14). Importantly, Armbruster et al. (13) identified a synthetic compound called clozapine-N-oxide (CNO) that is capable of activating muscarinic receptor–based DREADDs with high potency and efficacy. When used in the proper concentration or dose range, CNO is otherwise pharmacologically inert (13). More recently, novel DREADD agonists with improved pharmacodynamic and/or pharmacokinetic properties have been developed (15, 16). During the past years, Rosa-LSL-hM4Di mice, referred to as Rosa-LSL-GiD mice in this review, have been used to express the hM4Di (GiD) designer receptor in several metabolically important cell types including mouse pancreatic α-cells (17), hepatocytes (10), and adipocytes (11). CNO treatment of these mutant mouse strains leads to the selective stimulation of G_i signaling in specific cell types in vivo. This goal cannot be achieved by the use of receptor subtype–selective GPCR agonists targeting native GPCRs, since virtually all GPCRs are expressed in many different tissues (3).

Because of space constraints, this review cannot cover all metabolic effects that have been linked to G_i signaling in vivo or in vitro. Instead, this review will mostly focus on recent studies involving the use of Rosa-LSL-PTX and Rosa-LSL-GiD mice that have shed novel light onto the roles of G_i signaling in modulating blood glucose homeostasis in vivo.

Pancreatic β-Cells

The pancreatic islets of Langerhans contain, among other cell types, a large population of β-cells that are of key importance for maintaining proper blood glucose levels (euglycemia). β-Cells synthetize and release insulin, the key hormone involved in lowering elevated blood glucose levels. This process is regulated by numerous signaling pathways including the activity of β-cell G_i-coupled receptors (18).

Mice that selectively expressed the S1-PTX protein in pancreatic β-cells (RIP-PTX mice) showed pronounced metabolic phenotypes (9). RIP-PTX mice displayed elevated plasma insulin concentrations, which were accompanied by reduced blood glucose levels. Moreover, RIP-PTX mice showed a striking increase in glucose-stimulated insulin secretion (GSIS), greatly improved glucose tolerance, and were protected against the deficits in glucose homeostasis caused by the consumption of a high-fat diet (HFD) (Fig. 1) (9). In contrast, expression of a constitutively active G_i-coupled designer receptor (Ro1) in mouse β-cells during perinatal development resulted in reduced GSIS and impaired glucose tolerance in adult mice (Fig. 1) (19). These observations indicate that receptor-mediated activation of G_i signaling in β-cells leads to impaired insulin secretion and that drug-mediated inhibition of β-cell G_i signaling may prove useful to improve insulin release and glucose homeostasis under pathophysiological conditions.

Figure 1.

Modulation of G_i signaling in mouse pancreatic α- and β-cells affects whole body glucose homeostasis. The signaling pathways involved in the various metabolic phenotypes are described in detail in the main text. GSIS, glucose-stimulated insulin secretion.

Metabolic studies with mice transiently expressing the S1-PTX protein in β-cells during the perinatal period indicated that β-cell G_i signaling suppresses β-cell proliferation in a cell-autonomous manner, resulting in reduced β-cell mass and impaired glucose homeostasis in adult mice (Fig. 1) (19). Since a reduction of β-cell mass is a key feature of both type 1 (T1D) diabetes and type 2 diabetes (T2D) (20), this observation is of considerable translational interest.

β-Cells express many G_i-coupled receptors, although transcript levels vary greatly among the different receptors (18, 19). For example, glucose-stimulated somatostatin release from pancreatic δ-cells inhibits β-cell activity via paracrine activation of β-cell G_i-coupled somatostatin receptors (18). The G_i-coupled α _2A-adrenoceptor is expressed at particularly high levels by mouse and human β-cells (18, 19). Various lines of evidence suggest that activation of β-cell α _2A-adrenoceptors impairs insulin release from β-cells and that enhanced β-cell α _2A-adrenoceptor signaling can lead to impaired β-cell function in both mice and humans, leading to an increased risk of developing (T2D) (21). Somatostatin receptors and α _2A-adrenoceptors are expressed in many other cell types; it is therefore doubtful that blockers of these receptors can be developed as insulinomimetic drugs due to potential off-target side effects. Interestingly, a recent study demonstrated that activation of mouse β-cell cannabinoid 1 receptors (CB1Rs; a G_i-coupled receptor) inhibits β-cell function and mediates islet inflammation under conditions of metabolic stress (22). For this reason, β-cell CB1Rs represent a potential therapeutic target.

Pancreatic α-Cells

Besides β-cells, pancreatic islets also contain α-cells, δ-cells, and several other cell types. Pancreatic α-cells synthesize and release glucagon, a polypeptide hormone/paracrine factor that is critical for the maintenance of euglycemia (23-26). Impaired α-cell function is predicted to contribute to deficits in glucose homeostasis in T1D and T2D (26-29). Glucagon release from α-cells is strongly inhibited by somatostatin (SS) release from adjacent δ-cells (18). SS acts on G_i-coupled SS receptors (SSTR2 in mouse and SSTR1/SSTR2 in human) to reduce glucagon release by triggering the opening of inwardly rectifying K⁺ channels, thus suppressing electrical activity and directly inhibiting glucagon exocytosis (30).

To explore the physiological relevance of G_i-mediated inhibition of α-cells, Zhu et al. (17) recently generated a mouse model that expressed the GiD DREADD selectively in α-cells of mutant mice (α-GiD mice). Strikingly, acute CNO treatment of α-GiD mice led to pronounced reductions (by up to ~90%) in plasma glucagon levels, most likely due to G_i-mediated inhibition of glucagon release (17). The CNO-induced reduction in plasma glucagon levels was accompanied by a significant decrease in plasma insulin levels and impaired glucose tolerance, suggesting that efficient insulin release requires β-cell activation by glucagon (Fig. 1). Consistent with this notion, perifusion studies with isolated mouse and human islets have shown that efficient GSIS requires intra-islet glucagon that acts primarily on β-cell GLP-1 receptors (17). The finding that intra-islet glucagon represents a key paracrine factor capable of promoting GSIS is in agreement with several other recent reports (31-33). As reviewed by El et al. (34), intra-islet glucagon may have a similar role in human islets.

Taken together, these recent studies have led to a reassessment of the physiological roles of glucagon, suggesting that glucagon has dual, opposing actions on blood glucose levels, depending on the nutritional state of the organism. Traditionally, glucagon is known to act as a functional antagonist of insulin, due to its ability to raise blood glucose levels under fasting conditions, primarily by stimulating hepatic glucose production (HGP). However, in the fed state, as shown in a series of recent studies, intra-islet glucagon is required to allow glucose to stimulate insulin release with high efficacy. These dual actions of glucagon are highly relevant for the design of novel drugs targeting glucagon receptors or glucagon-regulated cellular pathways for therapeutic purposes.

Hepatocytes

Hepatocytes of the liver play a central role in regulating whole body glucose and lipid homeostasis and many other metabolic functions. In the diabetic state, glucose release from the liver is unphysiologically high, due to elevated plasma glucagon levels and the reduced sensitivity of hepatocytes to the effects of insulin and/or decreased plasma insulin levels (35). Hepatocytes store large amounts of glucose in the form of glycogen and are the main producers of glucose that is released into the blood. Because of the central role of the liver in the regulation of glucose homeostasis, much research has been focused on developing strategies to decrease HGP for therapeutic purposes (35).

Hepatocytes express a large number of GPCRs many of which are selectively linked to G proteins of the G_i family (3). To explore how activation of hepatocyte G_i signaling affects glucose homeostasis in vivo, Rossi et al. (10) generated mice that expressed the GiD designer receptor selectively in hepatocytes (hep-GiD mice). CNO treatment of hep-GiD mice led to pronounced increases in blood glucose levels in vivo, due to a sustained increase in the rates of both glycogenolysis and gluconeogenesis (Fig. 2). The G_i-induced increase in HGP did not require the presence of glucagon or hepatic glucagon receptors (10). Additional studies showed that G_i-mediated increases in HGP in vitro were absent in the presence of a selective inhibitor of c-Jun N-terminal Kinase (JNK) and that CNO-induced hyperglycemia in Hep-GiD mice was strongly reduced by the hepatocyte-specific expression of a dominant negative version of JNK. G_i-mediated activation of hepatic JNK activity increased the hepatic expression of glucose-6-phosphatase (G6Pase) and phosphoenolpyruvate carboxykinase, the 2 rate-controlling enzymes in the process of gluconeogenesis (10). Taken together, these data strongly suggest that G_i-mediated activation of JNK plays a key role in promoting HGP (Fig. 2).

Figure 2.

Activation or inhibition of G_i signaling in mouse hepatocytes causes major changes in whole body glucose homeostasis. The signaling pathways underlying the displayed metabolic phenotypes are described in detail in the main text. ROS, reactive oxygen species; JNK, c-jun N-terminal kinase, HGP, hepatic glucose production.

RNAseq studies indicated that activation of hepatocyte G_i signaling resulted in the upregulation of many genes associated with increased endoplasmic reticulum (ER) stress and the linked unfolded protein response (10). Previous studies have shown that ER stress and the resulting unfolded protein response promote the generation of reactive oxygen species (ROS) and JNK activation (36-38). Consistent with this observation, pretreatment of hep-GiD hepatocytes with N-acetyl cysteine, a ROS scavenger, completely blocked CNO-mediated increases in JNK phosphorylation (10). These findings support the existence of a signaling pathway through which hepatic G_i signaling stimulates ROS production, which in turn triggers the activation of JNK and enhanced HGP (Fig. 2).

Interestingly, human primary hepatocytes expressing a constitutively active version of Gα _i2 also showed a significant increase in glucose output (10). This effect was ROS-dependent and accompanied by significant increases in JNK phosphorylation and G6Pase expression, indicating that G_i signaling has similar effects on hepatic glucose fluxes in mouse and human hepatocytes.

Taking advantage of the availability of Rosa-LSL-PTX mice, Rossi et al. (10) recently generated mice that expressed the S1-PTX protein selectively in hepatocytes (hep-PTX mice). Interestingly, hep-PTX mice showed significantly improved glucose homeostasis, as compared to control littermates (Fig. 2). Importantly, when mice were maintained on an obesogenic diet, the metabolic deficits displayed by control mice (glucose intolerance, enhanced HGP, etc.) were absent in hep-PTX mice (10), indicating that the lack of hepatocyte G_i signaling has beneficial metabolic effects on whole body glucose homeostasis.

Hepatocytes express several endogenous G_i-coupled receptors including α ₂-adrenoceptors and CB1Rs (39, 40). In agreement with previous studies (39, 40), injection of wildtype mice with anandamide, a cannabinoid receptor agonist, resulted in impaired glucose tolerance (10). This metabolic deficit was absent in hep-PTX mice, indicative of the involvement of G_i-type G proteins. Additional studies showed that activation of endogenous hepatocyte CB₁ receptors promotes ROS formation, JNK activity, and HGP in a fashion similar to that observed with GiD-expressing hepatocytes (10). Studies with primary human hepatocytes strongly suggest that a similar signaling pathway is also operative in human liver (10).

In summary, the metabolic phenotypes displayed by hep-GiD and hep-PTX mice indicate that hepatic G_i signaling plays an important role in regulating hepatic glucose fluxes and whole body glucose homeostasis. Inhibition of hepatocyte G_i signaling by antagonists of G_i-coupled receptors (eg, peripherally acting CB1R antagonists) that are endogenously expressed by hepatocytes may prove useful to reduce elevated HGP for therapeutic purposes.

Adipocytes

Obesity has become the most common chronic disease worldwide, resulting in a dramatic increase in obesity-related metabolic disorders including T2D (41-43). Because lasting changes in lifestyle and diet are difficult to achieve, the development of novel drugs that are able to reduce appetite and/or stimulate energy expenditure represents a major goal of many research laboratories.

In obese individuals, adipocytes undergo hypertrophy and macrophage infiltration, leading to the release of inflammatory adipokines and excessive amounts of free fatty acids (FFAs) into the blood stream (41-43). These metabolic changes eventually reduce the ability of peripheral tissues to properly respond to insulin (peripheral insulin resistance), resulting in impaired glucose homeostasis.

Adipocyte function is modulated by a wide array of GPCRs including receptors that preferentially activate G proteins of the G_i family (3, 44). To study the effects resulting from the selective activation of G_i signaling in adipocytes in vivo, Wang et al. (11) generated and characterized a novel mouse strain that expressed the GiD DREADD selectively in adipocytes (adipo-GiD mice). CNO treatment of lean or obese adipo-GiD mice, but not of control littermates, led to marked reductions in plasma FFA, glycerol, and triglyceride levels, in agreement with previous studies indicating that G_i signaling exerts antilipolytic effects (45, 46). Activation of adipocyte G_i signaling also led to reduced plasma insulin levels, most likely due to increased insulin sensitivity resulting from lowered plasma FFA levels. Importantly, CNO-treated adipo-GiD mice displayed pronounced improvements in glucose tolerance and insulin sensitivity, particularly when mice were maintained on an obesogenic diet (Fig. 3) (11). These beneficial metabolic effects were accompanied by enhanced insulin-induced glucose uptake into adipose tissues. In a closely related, study, Caron et al. (47) reported that CNO treatment of adipo-GiD mice had no significant effect on glucose metabolism or lipolysis. The most likely explanation for the discrepant findings between the 2 studies is that Caron et al. used a 10-fold lower dose of CNO (1 mg/kg intraperitoneally) than Wang et al. (11).

Figure 3.

Stimulation or inhibition of G_i signaling in mouse adipocytes exerts pronounced effects on whole body glucose homeostasis. The molecular basis of the observed metabolic phenotypes is described in detail in the main text. FFA, free fatty acid.

Wang et al. (11) also analyzed a mouse model that expressed the S1-PTX protein selectively in adipocytes (adipo-PTX mice), thus disrupting adipocyte G_i signaling. Somewhat expectedly, adipo-PTX mice showed metabolic phenotypes that were opposite to those displayed by CNO-treated adipo-GiD mice. HFD adipo-PTX mice showed more severe deficits in glucose tolerance and insulin sensitivity than their HFD control littermates (Fig. 3) (11). Hyperinsulinemic–euglycemic clamp studies confirmed that inactivation of adipocyte G_i signaling causes severe insulin resistance. Additional studies (11) demonstrated that adipocyte G_i signaling is required for maintaining the ability of multiple peripheral tissues or organs to properly respond to insulin. In contrast to CNO-treated adipo-GiD mice, adipo-PTX mice showed significantly increased plasma FFA levels, most likely due to enhanced lipolysis (11). Reduction of these elevated FFA plasma levels with an inhibitor of hormone-sensitive lipase significantly improved glucose tolerance and insulin sensitivity in HFD adipo-PTX mice (11), indicating that increased plasma FFA levels play a central role in triggering the metabolic deficits caused by the absence of adipocyte G_i signaling.

HFD adipo-PTX mice also showed severe adipose tissue inflammation, a phenomenon closely linked to whole body insulin resistance (11). In agreement with this finding, plasma levels of resistin and several proinflammatory cytokines, such as interleukin-1β and tumor necrosis factor-α, were significantly increased in HFD adipo-PTX mice. Interestingly, the activity of adipocyte PTP1B, an enzyme known to interfere with insulin receptor autophosphorylation (48), was significantly increased in the absence of adipocyte G_i signaling. Pharmacological inhibition of PTP1B restored efficient insulin signaling in adipocytes lacking functional G_i proteins (11), indicating that G_i-mediated inhibition of PTP1B activity plays an important role in the proper responsiveness of adipocytes to insulin. This inhibitory G_i effect required G_i-mediated reductions in adipocyte cyclic adenosine monophosphate levels and reduced protein kinase A (PKA) activity.

These new data suggest that G_i-coupled receptors that are endogenously expressed by adipocytes represent potential drug targets for treating disorders in glucose and lipid homeostasis. It is well known that adipocytes express the G_i-coupled hydrocarboxylic acid (HCA) receptors 1 and 2 (HCA₁ and HCA₂, respectively; previous names: GPR81 and GPR109A, respectively). The HCA₁ and HCA₂ receptors are activated by lactate and nicotinic acid (endogenous ligand: 3-hydroxy-butyrate), respectively (45). Interestingly, Wang et al. (11) identified several G_i-coupled receptors that are upregulated in mice maintained on an obesogenic diet, including various orphan receptors and several chemokine, somatostatin, and purinergic receptor subtypes. Targeting these receptors by selective agonists may prove beneficial for the treatment of various metabolic disorders.

It should be noted that nicotinic acid (niacin) is an US Food and Drug Administration–approved drug to treat primary hyperlipidemia and hypertriglyceridemia. Until recently, it was assumed that the ability of nicotinic acid to inhibit lipolysis via activation of adipocyte HCA₂ receptors was responsible for its antiatherogenic activity (45). However, more recent evidence suggests that the antiatherogenic effects of nicotinic acid involve alternative sites of action (45). Nevertheless, HCA₂ receptor agonists, due to their ability to inhibit lipolysis and lower plasma FFA levels, are thought to have considerable potential to reduce insulin resistance in T2D (49).

Concluding Remarks

As summarized in this brief review, receptor-mediated modulation of G_i signaling plays a key role in maintaining glucose homeostasis by regulating the activity of multiple metabolically important cell types. The studies summarized in this review do not provide clear information regarding the relative contribution of Gα _i vs. Gβy signaling to the metabolic phenotypes observed with the different mutant mouse models. This topic remains to be addressed in future experiments. More importantly, additional studies are needed to explore in more detail to which extent the functionally important G_i signaling pathways identified in mice are also relevant for human physiology and pathophysiology. Such studies are likely to accelerate the development of novel drugs targeting G_i-mediated signaling for the treatment of various metabolic disorders including T2D and obesity. As a first step, human G_i-coupled receptors that are highly enriched in specific, metabolically relevant cell types need to be identified. Targeting of these receptors by highly selective agonists or antagonists appears to represent the most direct approach to modulate G_i signaling for therapeutic purposes.

Search strategies

I searched the PubMed and Google Scholar databases using combinations of the terms “G protein,” “G_i,” “G protein-coupled receptor,” “metabolism,” “in vivo,” “diabetes,” “glucose homeostasis,” “pertussis toxin” and “mouse.”

Glossary

Abbreviations

CB1R

cannabinoid 1 receptor

CNO

clozapine-N-oxide

DREADD

Designer Receptors Exclusively Activated by a Designer Drug

FFA

free fatty acid

GPCR

G protein–coupled receptor

GSIS

glucose-stimulated insulin secretion

HCA

hydrocarboxylic acid

HGP

hepatic glucose production

HFD

high-fat diet

JNK, c-Jun N-terminal Kinase; LSL

loxP-stop-loxP

PTP1B

protein tyrosine-phosphatase 1B

PTX

pertussis toxin

ROS

reactive oxygen species

SS

somatostatin

T1D

type 1 diabetes

T2D

type 2 diabetes

Additional Information

Disclosures: The author has nothing to declare.

Data Availability

Data sharing is not applicable to this article because no data sets were generated or analyzed during the present study.