A finer tuning of G-protein signaling through regulated control of RGS proteins

Abstract

Regulators of G-protein signaling (RGS) proteins are GTPase-activating proteins (GAP) for various Gα subunits of heterotrimeric G proteins. Through this mechanism, RGS proteins regulate the magnitude and duration of G-protein-coupled receptor signaling and are often referred to as fine tuners of G-protein signaling. Increasing evidence suggests that RGS proteins themselves are regulated through multiple mechanisms, which may provide an even finer tuning of G-protein signaling and crosstalk between G-protein-coupled receptors and other signaling pathways. This review summarizes the current data on the control of RGS function through regulated expression, intracellular localization, and covalent modification of RGS proteins, as related to cell function and the pathogenesis of diseases.

this article is part of a collection on G Protein Kinase A Signaling in Cardiovascular Physiology and Disease. Other articles appearing in this collection, as well as a full archive of all collections, can be found online at http://ajpheart.physiology.org/.

Introduction

In the mid-1990's, several investigators (34, 38, 39, 43, 90) identified a new class of proteins that regulated the function of heterotrimeric G proteins in a variety of model systems. Further studies (80, 185) demonstrated that these proteins accelerate the intrinsic GTPase activity of Gα subunits, thereby functioning as GTPase-activating proteins. This family is now collectively known as the regulator of G-protein signaling (RGS) proteins and has been divided into eight subfamilies based on sequence homology (188). All of the RGS family members express the evolutionary conserved RGS domain, which interacts with the GTP-bound Gα subunits and accelerates their GTPase activity, with variable degrees of selectivity among Gα_i, Gα_q/11, and Gα_12/13 subunits. In addition, many RGS proteins possess extensions of various lengths, often containing one or more functional domains other than the RGS box (188). These domains and the regions outside of the RGS box may facilitate interactions with noncanonical partners and mediate G-protein-independent functions, (2, 149) or may serve regulatory purposes as discussed in this review.

Although the biochemical functions of RGS proteins have been extensively studied, our understanding of the physiological functions of RGS proteins is relatively modest. In this emerging field, investigators have positioned RGS proteins as key modulators in the cardiovascular (57, 187, 199) and neurologic (168, 174, 175) systems, lymphocyte trafficking and inflammatory disorders (29, 42), and cancer biology (81). The potential importance of RGS proteins in disease has driven efforts to pharmacologically target RGS protein function (154, 155); however, this approach would be most effective when applied with a greater understanding of RGS regulation. Therefore, identifying the mechanisms that control RGS activity has become an increasing area of investigation.

This review provides a comprehensive summary of the existing data exploring the mechanisms by which RGS proteins are regulated. Common modes of control of RGS activity include regulation of expression, intracellular localization, and posttranslational modifications. Therefore, we have organized this review into these specified modules and provide detailed information on how these mechanisms regulate specific RGS proteins in Tables 1–4. Importantly, a given RGS protein is often regulated by several mechanisms and, vise versa, several RGS proteins can be regulated by one mechanism within the same cell, which is described throughout the text. Furthermore, the regulation of a given RGS protein may differ dependent on the cell type, indicating that the knowledge obtained on a particular RGS protein obtained in one cell type may not be applicable for the other cell type. While the quality of data in the cited (peer-reviewed) publications is quite variable, we invite the readers to ascertain the stringency of the studies with their own criteria. By highlighting our current knowledge of the regulation RGS proteins, we propose potential areas in which research is needed for understanding how the fine tuning of G-protein signaling occurs through regulation of RGS proteins in cells, animal models, and humans.

Table 1.

Regulated expression of RGS proteins

RGS	Stimuli/Disease Models	Expression	mRNA/Protein	Mechanism	Cells/Tissues	Ref. No.
SST2	Pheromone	↑1–6 h	Protein	(GPA1)	Yeast	(37, 39)
RGS1	Concanavalin A	↑1–4 h	mRNA	(Gi), (PKC)	Mononuclear cells	(67)
RGS1	SAC/phorbol	↑2–96	mRNA	(PKC)	B lymphocytes	(71)
RGS1	Anti-CD40 + anti-μ	↑4–48 h	mRNA		B lymphocytes	(151)
RGS1	LPS	↑4–48 h	mRNA, protein		Dendritic cells	(150)
		↑0.25–1 h	mRNA	TLR1/2/4/6/9	Macrophages	(132)
		↓12–24 h	mRNA	TLR1/2/4/6/9	Macrophages	(132)
RGS1	IFN-β	↑4–20 h	mRNA, protein	IFNAR2c receptor	Mononuclear cells	(173)
RGS1 in vivo	LPS/sepsis	↑2 h	mRNA		Porcine heart	(122)
RGS2	Concanavalin A	↑0.5–1.5 h	mRNA	(Gi), (Ca2+)	Mononuclear cells	(67)
RGS2	Carbachol	↑0.5–4 h	mRNA	(Gi/q), PKC	SH-SY5Y cells	(157, 158)
RGS2	Angiotensin II	↑0.5–4 h	mRNA, protein	(Gi/q), PKC, iPLA2, PKA	VSMC	(56, 101, 192, 193)
		↑1–12 h	mRNA	(Gi/q), Ca2+/calmodulin	H295R adrenal cells	(137)
		↑1–12 h	mRNA	(Gi/q)	Cardiac fibroblasts	(200)
RGS2	S1P	↑1–2 h	mRNA	(Gq)	VSMC	(64)
RGS2	Phenylephrine	↑0.5–3 h	mRNA, protein	(Gi/q)	Cardiomyocytes	(62, 204)
RGS2	ATP	↑2 h	mRNA	(Gi/q)	Osteoblasts	(142)
RGS2	Phorbol	↑1–2 h	mRNA	PKC	VSMC	(56)
		↑0.5–6 h	mRNA	(PKC)	Cardiomyocytes	(62, 204)
		↑2 h	mRNA	(PKC)	Osteoblasts	(142)
		↑1–3 h	mRNA	PKC	SH-SY5Y cells	(157)
RGS2	PGE1/2	↑2 h	mRNA	(Gs)	T cells	(6)
RGS2	Parathyroid hormones	↑0.5–6 h	mRNA	(Gs), cAMP, PKA	Osteoblasts	(108, 142, 169, 176)
	(PTH, PTHrP)	↑1–3 h	mRNA	(Gs), cAMP, PKA	UMR106 cells	(89, 176)
RGS2	Isoproterenol	↑1–3 h	mRNA, protein	(Gs, cAMP)	Cardiomyocytes	(62, 116)
RGS2	Thyrotropin	↑1–24 h	mRNA	(Gs)	Thyroid epithelial cells	(48)
RGS2	Cholera toxin	↑1.5 h	mRNA	Gs, cAMP, PKA	UMR106 cells	(89)
RGS2	Forskolin	↑3 h	mRNA, protein	cAMP, PKA	Cardiomyocytes	(62)
		↑4–24 h	mRNA	(cAMP)	UMR106 cells	(89)
		↑4–24 h	mRNA	(cAMP)	VSMC	(28)
		↑1–3 h	mRNA, protein	cAMP, PKA	Osteoblasts	(142, 176)
RGS2	PDGF, FGF	↑4–24 h	mRNA		VSMC	(28, 60)
RGS2	LPS	↓1–24 h	mRNA	TLR1/2/4/6/9	Macrophages	(132)
		↓1–1.5 h	mRNA	TLR4, PKC-η, PLD	Raw264.7 cells	(95)
RGS2	IL-2	↓2 h	mRNA	Reversed by cAMP	T cells	(6)
RGS2	IFNβ-1a	↑72 h	mRNA		Mononuclear leukocytes	(55)
RGS2	H2O2, peroxynitrite	↑1–4 h	mRNA, protein		1321N1 cells	(203)
RGS2	Heat shock	↑0.5–2 h	mRNA		SH-SY5Y cells	(159)
		↑1–3 h	protein		1321N1 cells	(203)
RGS2	Camptothecin, geldanamycin	↑1–4 h	mRNA	Cdk2	SH-SY5Y cells	(158)
RGS2	Ischemia/recovery	↑12–24 h	mRNA, protein	PKC	Astrocytes	(47)
RGS2	Mechanical stress	↑6–24 h	mRNA, protein	cAMP	Periodontal ligament cells	(147)
RGS2	Differentiation	↑0.5 h–9 days	mRNA	cAMP, SP1	3T3-L1 cells	(26, 82)
RGS2	Differentiation	↓1–5 days	mRNA		SH-SY5Y cells	(159)
RGS2 in vivo	PTH, PGE2	↑1 h	mRNA	(Gs), inhibited by vitamin D	Bone	(108, 178)
RGS2 in vivo	Angiotensin II	↑1 day	mRNA	(Gq)	Isolated cardiac fibroblasts	(200)
		↓3–14 days
RGS2 in vivo	Ovulation (hCG)	↑4–12 h	mRNA		Ovary	(179)
RGS2 in vivo	Cardiac hypertrophy:
	Gαq* transgenic mice	↓0.5–14 mo	mRNA, protein	Gαq	Heart, cardiac myocytes	(202)
	Aortic constriction	↓2–7 days	mRNA		Heart	(202)
RGS2 in vivo	Vascular hypertrophy	↓2 days	mRNA	Angiotensin II dependent	Aorta (abdominal, thoracic)	(182)
		↑14–28 days
RGS2 in vivo	Cardiac resynchronization	↑3 wk	Protein		Heart	(20, 21)
RGS3	Forskolin	↓4–24 h	mRNA	(PKA)	VSMC	(28)
RGS3	FGF	↑24 h	mRNA		Cardiomyocytes	(201)
RGS3 in vivo	Congestive heart failure	↓16 mo	mRNA, protein		Heart (SHHF rats)	(201)
RGS3 in vivo	Cardiac resynchronization	↑3 wk	Protein		Heart	(20, 21)
RGS4	Angiotensin II	↑2–12 h	mRNA	(Gq), CaM, CaMK, PKC	H295R cells	(138)
RGS4	Opioids	↑1–8 h	mRNA	Gi	PC12-μOR/κOR transfected cells	(112)
		↓18 h	Protein	Gi, proteasomal degradation	SH-SY5Y cells	(181)
RGS4	Adenosine	↓3 h	mRNA	cAMP	PC12
RGS4	Forskolin	↓3 h	mRNA	cAMP	PC12	(127)
		↓24 h	mRNA	(cAMP)	VSMC	(28)
RGS4	IL-1β	↑3 h-3 days	mRNA, protein	ERK, p38, NF-kβ	Colonic muscle strips and SMC	(76–78)
RGS4	Camptothecin, geldanamycin	↓1–4 h	mRNA	Cdk2	SH-SY5Y cells	(158)
RGS4	Proteasome inhibitors	↑4 h	Protein	RGS stabilization	Breast cancer cells	(190)
RGS4	FGF	↑24 h	mRNA		Cardiomyocytes	(201)
RGS4	PDGF-BB, PDGF-DD	↓6–24 h	mRNA		VSMC	(60)
RGS4	Dedifferentiation	↓Passages 2–6	mRNA, protein		VSMC	(66)
RGS4	Differentiation (NGF)	↓24 h	mRNA, protein		PC12 cells	(93)
RGS4	Differentiation	↑Day 15	mRNA		Chondrocytes	(3)
RGS4 in vivo	Cardiac hypertrophy	↑7 days	mRNA		Heart	(201)
RGS4 in vivo	Congestive heart failure	↓16 mo	mRNA, protein		Heart (SHHF rats)	(201)
RGS4 in vivo	Low-salt diet	↑15 days	mRNA		rat adrenal gland	(138)
	Angiotensin II infusion	↑2–6 h	mRNA
RGS4 in vivo	Vascular hypertrophy	↓2 days	mRNA	Angiotensin II dependent	Aorta (abdominal, thoracic)	(182)
		↑14–28 days
RGS4 in vivo	PPAR-δ (apoE−/− mice)	↑8 wk	mRNA		Atherosclerotic plaque	(5)
RGS4 in vivo	LPS treatment	↑24–72 h	mRNA, Protein		Heart	(124)
RGS5	β2AR-agonsists	↓4–24 h	mRNA	(Gαs)	Airway SMC	(198)
RGS5	Hypoxia	↑1–12 h	mRNA	HIF-1α	HUVEC	(88)
RGS5	PDGF-BB, PDGF-DD	↓6–24 h	mRNA		VSMC	(60)
RGS5	Dedifferentiation	↑Passages 2–5	mRNA		VSMC	(66)
RGS5	Differentiation	↓Days 6–15	mRNA		Chondrocytes	(3)
RGS5 in vivo	Atherogenic diet	↓6 mo	mRNA		Atherosclerotic plaque	(99)
RGS5 in vivo	PPAR-δ (apoE−/− mice)	↑8 wk	mRNA		Atherosclerotic plaque	(5)
RGS5 in vivo	Isoproterenol	↑14 days	mRNA, protein	β2AR mediated	Atria	(86)
RGS5 in vivo	Vascular hypertrophy	↓2 days	mRNA	Angiotensin II dependent	Aorta (abdominal, thoracic)	(182)
		↑14–28 days
RGS7	Differentiation	↑Days 6–15	mRNA		Chondrocytes	(3)
RGS7	TNF-α	↑4 h	Protein	p38MAPK, RGS7 stabilization	ECV-304 cells	(7)
	TNF-α, LPS	↑48 h	mRNA, protein		RAW cells, BV2 cells	(63)
RGS7 in vivo	TNF-α, LPS	↑4–6 h	Protein	TNFR1	Brain	(7)
	Spinal cord injury	↑48 h	Protein		Microglia/macrophages	(63)
RGS10	Cisplatin	↑24–48 h	mRNA		SKOV-3 cells	(72)
RGS10	Differentiation	↑Days 9–15	mRNA		Chondrocytes	(3)
RGS10	Tea polyphenols	↑4–24 h	mRNA		Colon cancer cells	(104)
RGS13	cAMP	↓1.5–6 h	mRNA	PKA	Mast cells	(194)
		⇆	Protein	RGS13 protein stabilization
RGS13	Anti-CD40 + anti-μ	↑4–48 h	mRNA		B lymphocytes	(151)
RGS13	IL-17	↑2–4 h	mRNA		B cells	(189)
RGS14	LPS	↓4–48 h	mRNA, protein		Dendritic cells	(150)
RGS14	Doxorubicin	↑16 h	mRNA	p53	RKO, MCF7, U-87, EB1	(15)
RGS14	Tea polyphenols	↑8–16 h	mRNA		Colon cancer cells	(104)
RGS16	PGE1/2	↓4–48 h	mRNA	(Gs)	T cells	(6)
RGS16	Endothelin-1	↑24 h	Protein	(Gq), Rho	Cardiac myocytes	(161)
RGS16	S1P	↑24 h	Protein	Gi, Rac	Cardiac myocytes	(161)
		↑1–2 h	mRNA	Gi	VSMC	(64)
RGS16	LPS	↑2 h	mRNA	NF-κB	Pre-B cells	(100)
		↑24 h	mRNA, protein	IL-1 secretion	Cardiac myocytes	(124, 125)
		↑2–6 h	mRNA		VSMC, aorta	(65)
		↑4–24 h	mRNA		Dendritic cells	(150)
RGS16	TNF-α	↑24 h	Protein	IL-1 secretion	Cardiac myocytes	(125)
		↑1–4 h	mRNA	Inhibited by Ca2+	T lymphocytes (CEM)	(50)
RGS16	IL-1	↑24 h	Protein		Cardiac myocytes	(125)
RGS16	IL-2	↑2 h	mRNA	Inhibited by cAMP	T lymphocytes	(6)
RGS16	IL-17	↑2–8 h	mRNA, protein	Traf3/6, NF-κB	B cells	(189)
RGS16	Phorbol	↑2–8 h	mRNA	PKC, TNFα secretion	T lymphocytes (CEM)	(50)
RGS16	Phitoemagglutinin	↑72 h	mRNA	Reversed by IFNβ-1a	Mononuclear leukocytes	(55)
RGS16	Differentiation	↑(N/D)	mRNA		Megakaryocytes	(12)
RGS16	Dedifferentiation	↓Passages 3–6	mRNA		VSMC	(66)
RGS16 in vivo	LPS/sepsis	↑2 h	mRNA		Heart (porcine)	(122, 124)
		↑24–72 h	protein		Heart (rat)
RGS17	Cisplatin	↑24–48 h	mRNA		SKOV-3 cells	(72)
RGS18	RANKL	↓4 days	mRNA, protein	(OGR1/GPR68, Gq)	RAW264.7, monocytes	(84)
RGS18	LPS	↓4–24 h	mRNA		Dendritic cells	(150)
RGS18	Differentiation	↑(N/D)	mRNA		Megakaryocytes	(12)
RGS19/GAIP	Differentiation	↓20–30 days	mRNA	Gi	HT-29, Caco-2 cells	(118)
RGS20/RGSZ	LPS	↑4 h	mRNA		Dendritic cells	(150)

Data are grouped for each regulator of G-protein signaling (RGS) protein based on stimuli as follows: G-protein-coupled receptor (GPCR) stimuli/signaling (G_q/i followed by G_s), cytokines, stress, growth/differentiation conditions, and in vivo models. GPCR agonists or drugs acting through G-protein signaling are in italics. Data are grouped for each regulators of RGS protein in chronological order of publications. The predicted, but not confirmed, signaling is in parentheses. Functional significance of regulated RGS expression is discussed in the review.

AR, adrenergic receptor; S1P, sphingosine-1-phosphate; HIF-1, hypoxia inducible factor; VSMC, vascular smooth muscle cells; HUVEC, human umbilical vein endothelial cells; PGE₂, prostaglandin E₂; PTH, parathyroid hormone; TLR, Toll-like receptors; PPAR, peroxisome proliferator-activated receptor; hCG, human chorionic gonadotropin; RANKL, receptor activator of NF-kB ligand; N/D, not described; ↑, up; ↓, down.

Table 2.

Regulation of RGS intracellular localization

RGS	Cell Type	Ectopic/Endogenous	Stimulus/Modification	Localization (From/To)	Ref. No.
RGS2	HEK293	Ectopic	Expression of active Gαq	Nucleus/plasma membrane	(68)
RGS2	HEK293	Ectopic	Expression of Gαs, Gαq, β2AR, AT1aR	Nucleus/plasma membrane	(141)
RGS2	COS-7	Ectopic	Heat shock, proteotoxic stress	Nucleus/nucleoli	(22)
RGS2	VSMC	Endogenous	Phosphorylation by PKGI	Cytoplasm/plasma membrane	(119, 167)
RGS2	PC3-AR	Ectopic	Melatonin, 8-bromo-cGMP	Nucleus/cytoplasm	(133)
RGS2	HEK293	Ectopic	Expression of active Gαq, M3R	Cytoplasm and nucleus/plasma membrane	(31)
RGS2	HEK	Ectopic	NH2-terminal 71 amino acids	Targets to plasma membrane	(58)
RGS2	M1-HEK	Ectopic	R44H (naturally occurring mutation)	Decreased affinity for plasma membrane	(59)
RGS3	HMC	Ectopic	Expression of active Gα11, AIF4−, NaF, endothelin-1	Cytoplasm/plasma membrane	(46)
RGS3	Rat astrocyte	Endogenous	Atrial natriuretic peptide	Cytoplasm/plasma membrane	(126)
RGS3	COS-7	Ectopic	Heat shock, proteotoxic stress	Nucleus/nucleoli	(22)
RGS4	S. cerevisiae	Ectopic	NH2-terminal 33 residues	Targets to plasma membrane	(10, 160)
RGS4	HEK293	Ectopic	Expression of active Gαi	Cytoplasm/plasma membrane	(44)
RGS4	Rat astrocyte	Endogenous	Atrial natriuretic peptide	Cytoplasm/plasma membrane	(126)
RGS4	HEK293	Ectopic	Expression of Gαi2, M2R	Cytoplasm/plasma membrane	(141)
RGS4	PC3-AR	Ectopic	Melatonin, 8-bromo-cGMP	Cytoplasm/nucleus	(133)
RGS4	HEK293	Ectopic	Expression of 14-3-3ε	Cytoplasmic retention	(1)
RGS4	SMC	Endogenous	Phosphorylation by PKA or PKG	Cytoplasm/plasma membrane	(79)
RGS6	COS-7	Ectopic	Heat shock, proteotoxic stress	Nucleus/nucleoli	(22)
RGS6	HEK293	Ectopic	Expression of R7BP	Cytoplasm/plasma membrane	(41)
RGS7	HEK-293	Ectopic	Gαo expression, palmitoylation	Cytoplasm/plasma membrane	(165)
RGS7	HEK293	Ectopic	Expression of R7BP	Cytoplasm/plasma membrane	(41)
RGS8	DDT1MF2	Ectopic	Expression of active Gαo, Gαi	Nucleus/plasma membrane	(106, 144)
RGS9-1	Sf9 cells	Ectopic	Expression of R9AP	Cytoplasm/plasma membrane	(74)
RGS9-1	HEK293	Ectopic	Expression of R7BP	Cytoplasm/plasma membrane	(41)
RGS9-2	HEK293	Ectopic	Expression of R7BP	Nucleus/plasma membrane	(41)
RGS10	HEK293	Ectopic	Phosphorylation by PKA (Ser168)	Plasma membrane/nucleus and cytoplasm	(17)
RGS10	PC3-AR	Ectopic	Melatnonin, 8-bromo-cGMP	Nucleus/cytoplasm	(133)
RGS11	HEK293	Ectopic	Expression of R7BP	Cytoplasm/plasma membrane	(41)
RGS13	HeLa	Ectopic	Expression of active Gαi	Cytoplasm/plasma membrane	(151)
RGS13	HeLa	Ectopic	Active Gαs expression	Cytoplasm/nucleus	(151)
	NIH3T3	Ectopic	cAMP, PKA expression	Cytoplasm/nucleus	(191)
RGS13	COS-7	Ectopic	Heat shock, proteotoxic stress	Nucleus/nucleoli	(22)
RGS14	HeLa	Ectopic	Mild heat stress	Cytoplasm/promyelocity leukemia bodies	(30)
RGS14	HeLa	Ectopic	Expression of Gαi1/3	Nucleus and cytoplasm/plasma membrane	(152)
RGS16	Yeast	Ectopic	NH2-terminal alpha-helix	Targets to plasma membrane	(23)
	COS-7	Ectopic	Palmitoylation	Targets to lipid rafts	(69)
RGS20	COS-7	Ectopic	Heat shock, proteotoxic stress	Nucleus/nucleoli	(22)
GAIP	COS-7	Ectopic	Palmitoylation	Cytoplasm/plasma membrane	(33)
p115RhoGEF	HEK293	Endogenous	U46619	Cytoplasm/plasma membrane	(14)
p115RhoGEF	MDCK, HeLa, U2-OS	Ectopic	Activation of G12/13 by S1P2	Cytoplasm /plasma membrane	(107)
p115RhoGEF	PC-12	Ectopic	Rho inhibition (C3)	Plasma membrane/cytoplasm	(13)

Data are grouped for each RGS protein in chronological order of publications.

Table 3.

Interaction of RGS proteins with GPCRs

RGS	In vitro/In Cells	Ectopic/Endogenous	Receptor/Site	Ref. No.
RGS2 (NH2 terminus)	In vitro, CHO	Ectopic	M1/3/5 muscarinic receptors (i3 loop)	(11)
RGS2	In vitro, HEK293	Ectopic	α-Adrenergic receptor (i3 loop, spinophilin)	(183)
RGS2	In vitro, CHO	Ectopic	α1Aadrenergic receptor (i3 loop: K219, S220, R238)	(61)
RGS2	In vitro, CHO	Ectopic	Sphingosine 1-phosphate receptor 1	(91)
RGS2 (NH2 terminus: K62, K63, Q67)	In vitro	Ectopic	Cholecystokinin receptor-2 (COOH terminus: P-S434, P-T439)	(94)
RGS4	In vitro, CHO	Ectopic	M1/5 muscarinic receptors (i3 loop)	(11)
RGS4	In vitro	Ectopic	μ/δ-Opioid receptors (i3 loop, COOH terminus)	(54)
RGS4	Rat striatum	Endogenous	Metabotropic glutamate receptor subtype 5	(148)
RGS4 (NH2 terminus)	In vitro, HEK293	Ectopic	μ/δ-Opioid receptors (i4 loop)	(97)
RGS7/Gβ5 (DEP domain)	In vitro	Ectopic	M3 muscarinic acetylcholine receptor (i3 loop, COOH terminus)	(146)
RGS8 (NH2 terminus, K8, K9)	In vitro, HEK293T	Ectopic	M1/M3 muscarinic acetylcholine receptors (i3 loop)	(83)
RGS8 (NH2 terminus)	In vitro, HEK293T	Ectopic	Melanin-concentrating hormone receptor 1	(109)
RGS9-2/ Gβ5	PC12	Ectopic, endogenous	μ-Opioid receptor	(130)
RGS9-2 (DEP domain)	PC12	Ectopic	D2 dopamine receptors	(92)
LARG (PDZ domain)	HEK293	Ectopic	LPA receptors 1 and 2 (COOH teminus)	(195)
PDZ-RhoGEF (PDZ domain)	HEK293	Ectopic	LPA receptor 1 and 2	(195)

Data are grouped for each RGS protein in chronological order of publications. The regions/residues and intermediate molecules providing the interaction are in parentheses.

Table 4.

Regulation of RGS proteins by covalent modification

RGS	Cell/Tissue	Endogenous/Ectopic	Covalent Modification	Site	Effect of Modification	Ref. No.
SstII	Yeast	Endogenous, ectopic	Phosphorylation	Ser539	Slows degradation of SstII	(52)
RGS2	HEK293T	Ectopic	Phosphorylation	P-tyrosine	Undetermined	(35)
RGS2	COS-7	Ectopic	Phosphorylation		Inhibits GAP activity	(32)
RGS2	VSMC, Rat1 fibroblasts	Endogenous	Phosphorylation	Ser46	Promotes membrane localization and GAP activity	(119, 167)
				Ser64
RGS2	In vitro	Ectopic	Palmitoylation	Cys116-	Inhibits GAP activity	(114)
				Cys106-	Promotes GAP activity
				Cys199-	Promotes GAP activity
RGS3	Astrocytes	Endogenous	Phosphorylation		Membrane localization	(126)
					Enhances RGS3 binding to Gαq and Gαi
RGS3	GGH(3)	Ectopic	Palmitoylation		Undetermined	(19)
RGS3	CHO	Ectopic	Phosphorylation	Ser-264	Provides interaction with 14-3-3	(115)
RGS4	Astrocytes	Endogenous, ectopic	Phosphorylation		Membrane localization	(126)
					Enhances RGS4 binding to Gαq and Gαi
RGS4	HEK293T	Ectopic	Phosphorylation	P-tyrosine	Undetermined	(35)
RGS4	Smooth muscle cells	Endogenous, ectopic	Phosphorylation	Ser52	Membrane localization	(79)
					Increases Gαq binding
RGS4	Cardiac myocytes	Endogenous	Phosphorylation		Increases interaction with Gαq	(171)
RGS4	S. cerevisiae	Ectopic	Palmitoylation	Cys2, Cys12	No effect on localization or activity	(160)
RGS4	Sf9	Ectopic	Palmitoylation	Cys2/12-	Promotes Cys95 palmitoylation	(177)
					inhibits or stimulates GAP activity (dependent on assay and Cys95 palmitoylation)
				Cys95-	Inhibits GAP activity
RGS4	HEK293	Ectopic	Palmitoylation	Cys95	Promotes GAP activity	(120)
RGS4	MDA-MB-231	Ectopic	Palmitoylation	Cys2	Stabilizes RGS5 protein	(180)
	HEK293				Blocks α-adrenergic Ca2+ signaling
RGS4	Embryonic fibroblasts	Ectopic	Arginylation	Cys2	Promotes degradation	(96)
RGS4	ATE1−/− mice, 3T3 cells	Endogenous, ectopic	Arginylation	NH2 terminus	Promotes degradation	(75)
RGS4	SH-SY5Y	Ectopic	Ubiquitination		Promotes degradation	(181)
RGS5	HEK293T	Ectopic	Phosphorylation	Ser166	Inhibits GAP activity	(110)
					Decreased Gα binding
RGS5	Embryonic fibroblasts	Ectopic	Arginylation	Cys2	Promotes degradation	(96)
RGS5	ATE1−/− mice, 3T3 cells	Endogenous, ectopic	Arginylation	NH2 terminus	Promotes degradation	(75)
RGS7	Mouse brain, HEK293T cells	Endogenous, ectopic	Phosphorylation	Ser241, Thr245, Thr247	Promotes protein stability	(7)
RGS7	Mouse brain, HEK293T cells	Endogenous, ectopic	Phosphorylation	Ser434	Decreases GAP activity by binding to 14-3-3	(9)
RGS7	Sf9	Ectopic	Palmitoylation		Membrane-bound RGS7 is palmitoylated	(139)
RGS7	HEK293	Ectopic	Palmitoylation	Cys133	Required for Gαo-mediated translocation to plasma membrane	(165)
RGS9-1	Outer rod, Sf9 cells, mouse retina	Endogenous, ectopic	Phosphorylation	Ser475	Phosphorylation is regulated by light	(73)
RGS9-1	Photoreceptor outer segments	Endogenous, ectopic	Phosphorylation	Ser427Ser428	Inhibits GAP activity	(4)
RGS9-1	Mouse retina, Sf9	Endogenous, ectopic	Phosphorylation	Ser475	Decreased affinity for R9AP	(156, 186)
RGS9-2	Periaqueductal gray matter membranes	Endogenous	Phosphorylation	Ser	Promotes 14-3-3 interaction	(53)
RGS10	HEK293	Ectopic	Phosphorylation	Ser168	Nuclear localization, sequestration from plasma membrane	(17)
RGS10	Platelets	Endogenous	Phosphorylation			(51)
RGS10	Sf9	Ectopic	Palmitoylation	Cys66	Inhibits or stimulates GAP activity (assay dependent)	(177)
RGS10	GGH(3)	Ectopic	Palmitoylation	Cys60	Required for inhibition of GnRH signaling	(19)
RGS13	HEK293T	Ectopic	Phosphorylation	Thr41	Inhibits degradation	(194)
RGS14	B35	Ectopic	Phosphorylation	Ser258, Thr494	Enhances GDI activity	(70)
RGS16	HEK293T	Ectopic	Phosphorylation	Ser53	Inhibits GAP activity	(24)
				Ser194
RGS16	HEK293T	Ectopic	Phosphorylation (EGFR)	Tyr168>Tyr177	Enhances GAP activity	(35)
RGS16	MCF-7, CHO-K1	Endogenous, ectopic	Phosphorylation (Src)	Tyr177>Tyr168	Enhances protein stability	(36)
RGS16	HEK293T	Ectopic	Palmitoylation	Cys2Cys12	Required for inhibition of Gi and Gq signaling	(45)
RGS16	HEK293	Ectopic	Palmitoylation	Cys98	Required for GAP activity	(120)
RGS16	HEK293/COS-7	Ectopic	Palmitoylation	Cys2Cys12Cys98	Required for inhibition of Gi/Gq signaling; lipid raft localization	(69)
RGS16	ATE1−/− mice, 3T3 cells	Endogenous, ectopic	Arginylation	NH2 terminus	Promotes degradation	(75)
RGS18	Platelets	Endogenous	Phosphorylation	Ser49		(51)
RGS19/GAIP	HT-29	Endogenous, ectopic	Phosphorylation	Ser151	Stimulates GAP activity	(117)
					Mediates autophagic pathway
RGS19/GAIP	AtT-20, HEK293T	Ectopic	Phosphorylation	Ser24		(49)
	Rat liver	Endogenous
RGS19/GAIP	COS7	Ectopic	Palmitoylation	NH2 terminus	Membrane-anchored GAIP is palmitoylated	(33)
RGS20/RGSZ	COS-7	Ectopic	Ubiquitination		Promotes degradation	(121)
RGS20/RGSZ	Periaqueductal gray matter	Endogenous	Sumoylation		Promotes Gαi and Gαz interaction	(134)
PDZ-RhoGEF	HEK293T	Ectopic	Phosphorylation	COOH terminus	Activation of Rho	(27)
LARG	HEK293T	Ectopic	Phosphorylation	Tyr	Promotes Rho activation by Gα12	(163)

Data are grouped for each RGS protein in the following order: phosphorylation, palmitoylation, and other modification.

GAP, GTPase-activating proteins; GnRH, gonadotropin-releasing hormone; GDI, guanine nucleotide dissociation inhibitor; GAIP, G alpha-interacting protein.

Regulated Expression of RGS Proteins

Table 1 summarizes the current data on the regulated expression of RGS proteins in response to numerous stimuli in various cell culture and animal models. Altogether, a few trends can be noted as described below.

Regulation of RGS expression by G-protein-coupled receptor signaling.

Early studies (37, 39) have demonstrated that the expression of yeast SST2, later recognized as a member of the RGS family, is upregulated in response to G-protein-mediated pheromone signaling. Subsequently, the expression of several RGS proteins has been shown to be increased upon stimulation of various cells with G_q/G_i-coupled agonists, including carbachol (RGS2), angiotensin II (RGS2 and RGS4), ATP (RGS2), phenylephrine (RGS2), sphingosine-1-phosphate (RGS2, RGS16), opioids (RGS4), and endothelin-1 (RGS16; see Table 1). The importance of G_q/G_i signaling in mediating agonist-induced expression of RGS proteins was demonstrated in many of these studies, potentially suggesting that stimulation of RGS expression by a given agonist may serve as a feedback mechanism for the regulation of its own signaling, as well as cross-desensitization of other G_q/G_i-coupled receptor signaling. For example, knockdown of RGS2, the expression of which is transiently (1–12 h) induced by angiotensin II in cultured cardiac fibroblasts, augmented angiotensin II-induced IP3 production, collagen synthesis, and proliferation (200). Interestingly, angiotensin II infusion in vivo resulted in a biphasic effect, with the initial induction of RGS2 expression in cardiac fibroblasts at day 1, followed by a decline below the control levels at days 3–14 (200). This may suggest that the initial RGS2 upregulation may serve to regulate angiotensin II signaling, whereas a subsequent RGS2 downregulation may contribute to the pathogenesis of cardiac hypertrophy and fibrosis; the latter is supported by other models (discussed in Regulation of RGS expression in animal models). Likewise, opioid receptor agonists induced a rapid but transient (1–8 h) induction of RGS4 mRNA in PC12-μOR/κOR transfected cells (112), whereas prolonged stimulation (24 h) of SH-SY5Y cells with μ- or δ- opioid receptor (MOR and DOR, respectively) agonists resulted in G_i-mediated ubiquitination and degradation of the RGS4 protein (181). Importantly, downregulation of RGS4 expression by MOR agonist (DAMGO) resulted in cross-sensitization of G_i/o-coupled DOR and the G_q-coupled M3 muscarinic receptor (M3R) signaling without affecting α₂-adrenergic receptor or bradykinin BK₂ receptor signaling (indicative of receptor-specific effects of RGS4; Ref. 181).

RGS2 expression is also stimulated in various cells by G_s-coupled agonists and cAMP-elevating drugs, including prostaglandin E₂ (6), parathyroid hormone (89, 108, 142, 169, 176), isoproterenol (62, 116), cholera toxin (89, 164), and forskolin (28, 62, 89, 123, 127, 142) (see Table 1). Together, these results may suggest that stimulation of RGS2 expression by a G_s-coupled agonist may provide cross-desensitization of G_q/G_i-coupled receptor signaling. However, only a few studies have directly assessed the possibility of cross-regulation between G_s and G_q/G_i by using knockdown/knockout approaches. In this context, studies by Roy et al. (142) demonstrated that forskolin pretreatment leads to desensitization of endothelin-1 and ATP signaling in wild-type (WT) but not in RGS2^−/− osteoblasts. RGS2 was shown to interact with various isoforms of adenylyl cyclases and to inhibit their activity (140, 141, 145, 153). Ectopic RGS2 overexpression has led to inhibition of cAMP production induced by odorant in olfactory neurons (153) or by isoproterenol in HEK293/β₂AR cells (140, 145) and in rat neonatal ventricular cardiomyocytes (116). This may suggest that stimulation of RGS2 expression by a G_s-coupled agonist may also lead to cross-desensitization of other G_s-coupled receptors through inhibition of adenylyl cyclase, which has yet to be demonstrated. Of note, other investigators (62) showed that RGS2 overexpression does not affect isoproterenol-induced cAMP production in rat adult ventricular myocytes. Given that RGS2 may have variable effects on different adenylyl cyclase isoforms, this effect of RGS2 may be cell specific, depending on which adenylyl cyclase isoforms are expressed in a given cell, and may differ even between adult (62) and neonatal (116) cardiomyocytes.

In contrast to RGS2, the expression of RGS3 (28), RGS4 (28, 112), RGS5 (198), and RGS16 (6) is downregulated by G_s/cAMP-coupled agonists in various cell types (see Table 1). Of interest is the regulation of RGS13 expression in mast cells, in which RGS13 mRNA is decreased by cAMP, whereas RGS13 protein is stabilized by PKA-dependent phosphorylation (194), suggesting that the balance between these two processes may determine RGS13 expression levels. cAMP is known to regulate the acute G_q/G_i signaling (i.e., Ca²⁺ release from intracellular stores) at least in part through 1) PKA-dependent phosphorylation and inhibition of phospholipase-Cβ₂ that produces IP₃ (103) or of IP₃ receptors (16); and 2) through activation of Ca²⁺ reuptake by sarcoplasmic reticulum Ca²⁺-ATPase (SERCA2) via PKA-dependent phosphorylation of the SERCA2 inhibitor phospholamban and subsequent release of phospholamban from SERCA2 (105). However, downregulation of RGS3/4/5/16 by Gs/cAMP-coupled agonists may suggest that G_q/G_i signaling could be also enhanced by cAMP over time. One study (198) addressed this functionally in airway smooth muscle cells and showed that downregulation of G_q/G_i-specific RGS5 mRNA and protein by long-term treatment with isoproterenol paralleled increased Ca²⁺ responses to bradykinin, thrombin, and histamine. Furthermore, RGS5 knockdown produced a similar effect in this study. However, as a given cell may express both RGS2 (that is induced by cAMP) and RGS3/4/5/16 (that are downregulated by cAMP), the overall effect of long-term treatment with G_s/cAMP-coupled agonists on G_q/G_i signaling may be cell specific, depending on the expression profile of RGS proteins.

Regulation of RGS expression by LPS and inflammatory cytokines.

Bacterial LPS and proinflammatory cytokines are important mediators of the immune response and inflammation. Not surprisingly, most of these studies were performed on immune cells, although some explored smooth muscle cells and cardiac myocytes, which are also targets of inflammation. In immune cells, LPS variably affected RGS expression, depending on the RGS protein and the cell type (see Table 1). LPS was shown to upregulate RGS1 in dendritic cells and macrophages (132, 150), RGS16 in pre-B and dendritic cells (100, 150), and RGS20/RGSZ1 in dendritic cells (150). Interestingly, in macrophages, the initial rapid upregulation of RGS1 mRNA by LPS was followed by a decrease in its expression below control levels (95, 132) . On the other hand, LPS decreased RGS2 mRNA in macrophages (95, 132), as well as RGS14 and RGS18 mRNAs in dendritic cells (150). Functionally, one study (95) using RGS2 small interfering siRNA and RGS2 overexpression approaches has demonstrated that downregulation of RGS2 mRNA by LPS in macrophages may mediate LPS-induced foam cell formation, although it remains to be determined whether this function of RGS2 relates to the regulation of G-protein signaling. The expression of some RGS proteins in various immune cells is induced by inflammatory cytokines, including interferon-β [RGS1 (173), RGS2 (55)], interleukin-1 [RGS16 (100)], interleukin-2 [RGS16 (6)], interleukin-17 [RGS13, RGS16 (189)], and TNFα [RGS16 (50)]. Given the established role of G_i signaling in migration of immune cells towards various chemokines, one would anticipate that the regulated expression of RGS proteins by LPS and inflammatory cytokines would affect this process, although this remains to be investigated.

In nonimmune cells, IL-1β has been shown to stimulate a rapid and sustained expression of RGS4 in colonic smooth muscle cells and in ex vivo-treated colonic muscle strips, which coincided with inhibition of acetylcholine-stimulated phospholipase C (PLC) activation and smooth muscle contraction by IL-1β (76–78); these effects were reversed by RGS4 knockdown (78). LPS, IL-1, and TNF-α increased the expression of RGS4 and RGS16 in cardiac myocytes, which corresponded with the inhibition of endothelin-1-induced PLC activation (124, 125). Finally, LPS stimulated expression of RGS16 in vascular smooth muscle cells and in ex vivo-treated aortic rings, which paralleled the inhibition of phenylephrine- and angiotensin II-induced contraction (65). Overall, these data suggest that RGS4/16 upregulation may diminish G_q/G_i signaling and contraction in LPS/TNFα-pretreated vascular smooth muscle cells and cardiac myocytes. Although these RGS proteins may contribute to the vasodilation and acute heart failure observed during sepsis, the causative role of RGS4 and RGS16 in the effects of LPS described in the latter studies (65, 124, 125) remains to be determined. Interestingly, while LPS-induced RGS16 expression was associated with inhibition of phenylephrine- or angiotensin II-induced contraction of ex vivo aortic rings, LPS pretreatment did not affect contraction in response to endothelin-1 and even enhanced serotonin-induced contraction (65). The former observation (related to endothelin-1) may represent a receptor-selective function of RGS proteins as reviewed previously (2, 113), although the interaction of RGS16 with G-protein-coupled receptors (GPCRs) has not been investigated (see Table 3). The latter observation (related to serotonin) may represent the regulation of smooth muscle contraction through a transporter-mediated serotonin uptake independent of G-protein signaling (102).

RGS expression under cell stress conditions.

Mammalian cells experience and variably respond to stress under various injurious conditions (hyperoxia, hypoxia/ischemia, mechanical stress and drug treatment). Several studies have examined how inducers of cell stress regulate RGS expression. RGS2 was shown to be upregulated in astrocytoma cells upon oxidative stress and heat shock or ischemia (47, 203) and in neuroblastoma SH-SY5Y cells upon heat shock and cell cycle arrest (157, 158). In the study of Endale et al. (47), overexpression of RGS2 mimicked ischemia-induced apoptosis. Similarly, hypoxia increased RGS5 expression in endothelial cells in a HIF-1α-dependent manner, which mediated hypoxia-induced apoptosis, as shown by utilizing RGS5 siRNA (88). Mechanical stress (compressive force) of periodontal ligament cells resulted in a cAMP-dependent expression of RGS2, which limited further cAMP production as determined using antisense S-oligonucleotides to RGS2 (147). Finally, treatment of ovarian cancer cells with therapeutic cytotoxic drugs (cisplatin, vincristine, and docetaxel) led to upregulation of RGS10 and RGS17, which mediated cell toxicity as determined by siRNA approach (72). These data suggest that increased expression of RGS proteins can mediate the detrimental effects of cell stress under certain conditions. Future studies should examine if and how such regulations of RGS expression are related to the control of G-protein signaling and function in the context of cell stress.

Regulation of RGS expression in animal models.

A large body of data on the regulation of RGS expression in the brain as related to neurological diseases and drug abuse was reviewed elsewhere (174, 175). This chapter focuses on the regulation of RGS expression in animal models as related to cardiovascular diseases, including hypertension, atherosclerosis, heart failure, and sepsis. Some of these topics are also reviewed elsewhere (155, 199).

It was initially reported that levels of RGS4 are increased in the pulmonary artery-banded mouse model of cardiac hypertrophy (201). In the follow-up study, this group showed that unlike WT mice, transgenic mice with cardiac-specific overexpression of RGS4 did not respond to phenylephrine by MAP kinase activation in the heart and exhibited reduced cardiac hypertrophy, with no induction of the “fetal” gene program, 7 days post-transverse aortic constriction (135). Interestingly, RGS4-transgenic mice had a significantly reduced survival 1–2 days post-transverse aortic constriction, suggesting that RGS4 also controls cardiac responses to acute pressure overload (135). Furthermore, when RGS4-transgenic mice were crossed with mice overexpressing Gα_q in the heart (a model for cardiac hypertrophy), the Gα_q/RGS4 transgenic mice had reduced cardiac hypertrophy and normalized left ventricular function, which was accompanied with reduced activity of PKCε and decreased expression of some hypertrophic genes (atrial natriuretic factor, skeletal muscle α-actin), compared with Gα_q-transgenic mice (136). In contrast to the above studies, the expression of RGS3 and RGS4 mRNA and protein was dramatically decreased in the heart during congestive heart failure in 16- to 20-mo-old SHHF rats (201). However, it would be interesting to examine the expression of these proteins at the earlier age (3–5 mo), when elevated blood pressure and cardiac hypertrophy are already developed before heart failure.

In contrast to RGS4, the expression of RGS2 was markedly decreased in three models of cardiac hypertrophy: transgenic cardiac-specific expression of an activated Gα_q mutant (202), aortic banding-induced pressure overload (202), and prolonged infusion of angiotensin II (200). Importantly, knockdown of RGS2 increased G_q signaling and hypertrophy of cultured cardiac myocytes and fibroblasts in these studies. The functional significance of RGS2 downregulation in mediating cardiac hypertrophy was further demonstrated using RGS2-knockout mice, which exhibit increased myocardial Gq signaling and cardiac hypertrophy, compared with WT mice (166). Furthermore, in the dyssynchronous heart failure canine model, cardiac resynchronization therapy resulted in increased cardiac expression of RGS2 and RGS3, which was associated with improved heart function and with suppressed Gα_i signaling, yielding increased Gα_s-biased cAMP responses to β₂-adrenergic stimulation in isolated cardiac resynchronization therapy myocytes. (20, 21). On the other hand, RGS4 and RGS16 mRNA and protein are increased in the heart in the LPS-induced model of sepsis known to be associated with acute heart failure, which coincided with decreased PLC activity in the heart (124).

In the abdominal or thoracic aorta banding models of hypertension and vascular hypertrophy, biphasic regulation of RGS2, RGS4, and RGS5 was observed in the thoracic and abdominal aortas, respectively. Banding resulted in a dramatic decline of their mRNAs on day 2, followed by a several fold increase over control aortas on day 28 postbanding (182), which is in contrast to a sustained downregulation of cardiac RGS2 in the thoracic aorta-banding model (202). This dynamic expression of RGS2/4/5 in the aorta correlated with increased G_q/G_i agonist-induced contraction of aortic rings on day 2 and decreased contraction on day 28 ex vivo (182). However, this study has not investigated how increased RGS2/4/5 may have contributed to the development of hypertension and vascular hypertrophy occurring during days 14–28. Given that RGS2^−/− mice develop enhanced vascular contraction and hypertension (167), it is possible that at least RGS2 upregulation plays a compensatory role in the former study. Additionally, RGS5 mRNA was sharply downregulated in atherosclerotic plaques in the atherogenic diet-induced model of atherosclerosis (99). In agreement with this study, the reduction of atherosclerosis by peroxisome proliferator-activated receptor-δ agonists in apoE^−/− mice was accompanied by an increase of RGS4 and RGS5 mRNAs in atherosclerotic plaques (5). These data are consistent with the proliferative phenotype of VSMC in plaques, which could be regulated by RGS4 and RGS5, although this notion is yet to be investigated.

Together, these data demonstrate that the expression of a given RGS protein is variably regulated in different models of cardiovascular disease, and the expression of more than one RGS protein may be altered in a given model. However, the specific contribution of each RGS protein to the development of disease in these models remains largely uncertain and requires further investigation.

Regulation of Intracellular Localization of RGS Proteins

To perform the canonical GAP function, RGS proteins must localize to the plasma membrane (where GPCRs and G proteins are localized). Several mechanisms for the regulation of RGS intracellular localization are summarized in Table 2 and are described below.

RGS2, RGS4, and RGS16 are targeted to the membrane by specific regions NH₂-terminal to their RGS domains (10, 23, 58, 160). The R44H mutation of RGS2 within the amino terminal amphipathic α-helix (responsible for plasma membrane-targeting) leads to less efficient membrane binding and inhibition of G_q signaling (59). Interestingly, this R44H mutation of RGS2 is associated with hypertension in the Japanese population (197). Palmitoylation of RGS proteins is another mechanism for membrane targeting, which is reviewed below (see Regulation of RGS Proteins by Covalent Modification).

RGS proteins can be also recruited to the membrane by activated (GTP-bound) Gα subunits that have a high binding affinity for RGS proteins. Many studies used overexpressed constitutively active (GTPase-deficient) Gα subunits to show translocation of RGS proteins from the cytosol to the membrane. Expression of constitutively active Gα_q/11 and/or Gα_i/o has led to membrane localization of RGS2 (31, 68), RGS3 (46), RGS4 (44), and RGS8 (106, 144). Given the transient nature of the RGS/Gα-GTP interaction, it is challenging to demonstrate GPCR agonist-induced RGS translocation to the membrane in intact cells. Nevertheless, membrane translocation of RGS3 in response to endothelin-1 (46), of RGS4 and RGS9–2 in response to μ/δ-opioid receptor agonists (97, 130), of p115RhoGEF in response to sphingosine-1-phosphate (S1P) or the thromboxane A(2) receptor agonist (13, 107), or of RGS3 and RGS4 in response to a non-G-protein coupled agonist, atrial natriuretic peptide (126), have been reported. The latter data suggest that the Gα-GTP/RGS interaction is not the only mechanism of RGS recruitment to the membrane and downstream signaling could be also involved. Indeed, roles for Ca²⁺ mobilization [RGS3 (46, 172)] and RhoA and PI3K activation [p115RhoGEF (13, 14)], or cGMP production [RGS2 (119, 133, 167), RGS3, RGS4 (126), and RGS10 (133)] have been suggested in driving membrane translocation of RGS proteins. Interestingly, some studies have shown that the expression of WT Gα subunits also promote membrane localization of RGS proteins, which includes the recruitment of RGS2 and RGS4 to the membrane by expression of WT Gα_q and Gα_s (141) and of RGS14 by expression of WT Gα_i1 and Gα_i3 (152). However, it is possible that overexpression of WT Gα subunits and RGS proteins may enable a low affinity interaction, which could be sufficient for the recruitment of RGS proteins to the membrane (our unpublished observations). Finally, membrane targeting of R7 family members (RGS6/7/9/11) is mediated through their interaction with membrane anchors, RGS7-binding protein (R7BP) and/or RGS9-anchoring protein (R9AP), which facilitate the regulatory function of R7 proteins, which is reviewed elsewhere (85). Even though the expression (and function) of R7 members of RGS family is largely restricted to the brain and retina, it was also shown that RGS6 is expressed in cardiac myocytes (40) and modulates parasympathetic regulation of heart rate through the regulation of muscarinic receptor-activated potassium channels (129, 196). Hence, modulation of RGS6 intracellular localization in cardiac myocytes may be of importance for heart function and disease, and should be evaluated in the future.

Several RGS proteins were shown to interact with various GPCRs (through regions outside of the RGS domain) by binding the third intracellular (i3) loop or the COOH terminus of GPCRs, either directly or through intermediate proteins such as spinophilin, which has been previously reviewed (2, 113). Table 3 provides updated information on RGS-GPCR interactions and the regions involved. While this interaction could provide receptor-specific selectivity of RGS proteins in regulation of signaling, it may also serve to recruit RGS proteins to the membrane. Indeed, RGS2 plasma membrane localization has been shown to be induced by expression of the β₂-adrenergic receptor (141), AT1A angiotensin II receptor (141), and the M3 muscarinic receptor (31), whereas RGS4 membrane localization was promoted by expression of the M2 muscarinic receptor (141) (Table 2). The COOH terminus and the i3 loop of GPCRs are known to be phosphorylated by GRK and PKA/PKC respectively, which mediates GPCR desensitization and internalization (170). Therefore, it would be interesting to examine if and how these processes are affected by RGS interaction and how the latter is affected by GPCR phosphorylation.

Finally, many RGS proteins were shown to undergo regulated localization to intracellular compartments other than the plasma membrane, including the nucleus [reviewed by Sethakorn et al. (149)]. While nuclear localization of RGS proteins may suggest functions unrelated to regulation of G-protein signaling, it may also serve to sequester RGS proteins from the cytoplasm, thus regulating their canonical function. Interestingly, some studies suggest that nuclear/cytosolic localization is a regulated process. For example, melatonin receptor activation led to nuclear accumulation of RGS4 but induced translocation of RGS2 and RGS10 from the nucleus to the cytoplasm (133), whereas PKA-dependent phosphorylation of RGS10 induced its nuclear translocation and thus nullified RGS10 activity at the plasma membrane (17).

Regulation of RGS Proteins Through Covalent Modifications

Table 4 summarizes the current data on the effect of posttranslational modifications on the activity of RGS proteins. Prominent covalent modifications include phosphorylation and palmitoylation; however, ubiquitination, sumoylation, and arginylation have also been reported. These modifications have a prominent role in modulating RGS protein function and also represent mechanisms of G-protein signaling regulation by both G-protein and non-G-protein pathways.

Regulation of RGS proteins by phosphorylation.

Phosphorylation of many RGS proteins has been reported, spanning across all RGS subfamilies. In several cases, phosphorylation of RGS proteins [RGS2 (32), RGS5 (110), RGS9–1 (4), and RGS16 (24)] resulted in inhibition of GAP activity, as demonstrated by in vitro GTPase activity assays and measurements of downstream signaling effectors. Alternatively, phosphorylation can indirectly inhibit the GAP activity of RGS proteins. For example, phosphorylation of RGS3 outside of the RGS domain (115, 184) and of RGS7 within the RGS domain (9) leads to their interaction with the phospho-serine-binding protein 14–3-3, resulting in inhibition of RGS activity. Structural studies revealed that 14–3-3 binding induces conformational changes in both the NH₂-terminal region and the COOH-terminal RGS domain of phosphorylated RGS3 (131), while functional studies demonstrated that 14–3-3 binding to RGS3 inhibits its interaction with G proteins (115). Similarly, RGS4, RGS5, and RGS16 were reported to interact with 14–3-3, which led to inhibition of RGS4 and RGS16 GAP activity through competition between 14–3-3 and the RGS/G-protein interaction (1). However, RGS5 GAP function was not significantly affected, which may indicate that the 14–3-3 interaction as a mechanism of regulation may be specific to certain RGS proteins (1).

Conversely, phosphorylation of RGS4 in astrocytes (126), smooth muscle cells (79), and cardiac myocytes (171) appears to stimulate their GAP activity through increased interaction with Gα_q and/or membrane localization. In smooth muscle cells, inhibition of PLC activity by cAMP was mediated by PKA-dependent phosphorylation of GRK2 [containing a G_q-specific RGS-like domain(18)] and of RGS4 (79). Overall, these results suggest that the effects of phosphorylation on RGS function may be cell-type specific, may differ depending on whether proteins tested were endogenous or ectopic, or may be dependent on the site of phosphorylation and/or kinase. In support of the latter notion, phosphorylation of RGS2 by PKG (at S46/S64) increased RGS2 GAP activity (167), whereas phosphorylation by PKC (site unknown) inhibited RGS2 GAP activity (32). Likewise, phosphorylation of RGS16 at S53/S194 inhibits GAP activity (24), whereas phosphorylation at Y168/Y177 by the EGF receptor or by Src kinases augmented RGS16 inhibition of G_i signaling, either through enhanced GAP activity (35) or through RGS16 protein stabilization (36). Of note, RGS2 and RGS4 can be also phosphorylated on tyrosine residues (35), although the specific sites and the significance of phosphorylation remain to be determined.

As described above for RGS16 (36), phosphorylation can promote the stability of RGS proteins. It was first demonstrated that pheromone-induced phosphorylation of SST2 at S539 by MAP kinase stabilized the SST2 protein in yeast (52). Furthermore, Xie et al. (194) reported that while cAMP significantly reduced mRNA levels of RGS13, the protein levels remained largely constant due to PKA-mediated phosphorylation and subsequent stabilization of the RGS13 protein, likely through protection from proteasomal degradation (194). Functionally, this promoted RGS13-mediated inhibition of carbachol-induced ERK phosphorylation (194). RGS7 provides another example of phosphorylation-regulated stability. Benzing et al. (7) reported that TNF-α treatment promoted the stability of RGS7, which was mediated by p38MAP kinase-dependent phosphorylation of RGS7 at S241, T245, and T247 sites. These authors also demonstrated that in vivo administration of TNF-α or LPS resulted in increased RGS7 protein expression in mouse brain (7). In addition to promoting long-term protein stability, TNF-α treatment also resulted in short-term dephosphorylation of RGS7 at S434 and concomitant loss of interaction with 14–3-3, which negatively regulates the activity of RGS7 (8). Thus RGS7 exemplifies a protein that can be regulated by the same stimulus through dual, phospho-site-specific, mechanisms. However, it is noteworthy that the major mechanism for stabilization of RGS7 as well of other members of the R7 family (RSG6/9/11) is likely through their interaction with Gβ5, as knockout of Gβ5 results in a drastic reduction or elimination of R7 protein expression without having an effect on mRNA levels (25).

Protein phosphorylation has been also demonstrated to affect localization of RGS proteins. Pedram et al. (126) reported that atrial natriuretic peptide induced PKG-mediated phosphorylation of RGS3 and RGS4, enhancing membrane localization and association with Gα_q and Gα_i and presumably augmenting G_q/G_i regulation. On the other hand, 14–3-3 interaction with phosphorylated RGS3/4/5/7/16 proteins (1, 9, 115) may also affect their localization through cytoplasmic retention by 14–3-3, as was demonstrated for 14–3-3-bound RGS4 (1). Phosphorylation of RGS10 by PKA induced its translocation to the nucleus, thus preventing its GAP function at the plasma membrane (17). Cyclic AMP/PKA signaling also induced nuclear translocation of RGS13, although indirectly through association of RGS13 with PKA-phosphorylated CREB (191). Interestingly, in the nucleus, RGS13 acts also noncanonically as a repressor of CREB-dependent gene transcription (191).

Regulation of RGS proteins by palmitoylation.

Palmitoylation of cysteines has been repeatedly reported to promote membrane localization, as is the case for several RGS proteins including RGS4 (120, 128, 177, 180), RGS7 (139, 165), RGS16 (45, 69, 120), and GAIP (33). Logically, these palmitoylated RGS proteins would demonstrate increased GAP activity secondary to increased colocalization with their G-protein partners at the membrane. However, data from several groups suggest that the effects of palmitoylation on RGS proteins may be more complex. For example, Ni et al. (114) demonstrated that the specific palmitoylated residues on RGS2 dictate whether the modification inhibits (Cys-116) or promotes (Cys-106, Cys-199) GAP activity, at least in vitro. Palmitoylation of RGS4 within the RGS box at Cys-95 inhibited GAP activity in Sf9 cells (177) but promoted it in HEK293 cells (120). Palmitoylation of RGS4 outside of the RGS box at Cys-2 and Cys-12 promoted palmitoylation at Cys-95 and might also modulate RGS4 GAP activity, although the effect varied dependent on the GAP assay (177). Palmitoylation of RGS4 at Cys-2 also increased RGS4 stability and potentiated the inhibition of α-adrenergic-induced calcium signaling by RGS4 (180). Other studies suggest that specific membrane compartmentalization, compared with general membrane binding, may have a greater effect of RGS function, at least for some RGS proteins. Mutation of cysteine residues crucial for palmitoylation in RGS16 demonstrated that while membrane binding was not affected, localization into lipid rafts was severely abrogated and resulted in decreased GAP activity of RGS16 (45, 69, 120). Palmitoylation of RGS10 could either attenuate or potentiate GAP activity, depending on whether the assay was performed as an in-solution single-turnover assay or in reconstituted phospholipid vesicles measuring steady-state GTP hydrolysis (177). However, in GGH3 cells, expression of a palmitoylation-deficient mutant of RGS10 attenuated its ability to regulate GnRHR-mediated PLC activation (19), which is more consistent with a positive role for palmitoylation on RGS10 GAP activity. In the same study, the authors (19) also reported that palmitoylation of RGS3 was dependent on GnRHR stimulation, although a functional significance was not investigated.

Regulation of RGS proteins by arginylation, ubiquitination, or sumoylation.

Arginylation, the process of conjugating arginine residues to NH₂-terminal aspartate, glutamate, and cysteine residues, is a key step in the ubiquitin-dependent proteasomal degradation N-end rule pathway (143). Several RGS proteins were demonstrated to be arginylated, resulting in the degradation of these proteins. Hu et al. (75) reported that the arginyl-transferase ATE1 could modify RGS4/5/6, and expression of these proteins was dramatically increased in ATE1^−/− mice. Of note, increased expression of these RGS proteins was not observed in all tissues of the knockout animals, but rather RGS4 and RGS16 appeared to be strongly upregulated in the brain and lung, whereas RGS5 showed robust stabilization in the heart (75). These authors also demonstrated that RGS4 was stabilized under low oxygen conditions, and a concurrent study (96) showed that proteolysis of RGS5 was decreased during hypoxia. The increased levels of RGS proteins in ATE1 knockout mice are postulated to contribute to the massive cardiovascular defects observed in these animals, and indeed ERK activation and induction of GPCR-regulated genes are impaired in the ATE1-deficient embryos (96). Overall these data suggest an intricate interplay of oxygen-sensing mechanisms and regulation of RGS protein levels to finely tune G-protein activity in the heart and possibly in other tissues as well.

Other groups have examined receptor-mediated downregulation of RGS proteins. For example, in neuroblastoma SH-SY5Y cells, μ- and δ-opioid receptor (MOR and DOR) activation led to ubiquitination and proteolysis of RGS4. This degradation was sensitive to pertussis toxin and proteasomal and lysosomal inhibitors. Furthermore, downregulation of RGS4 by a MOR agonist resulted in augmentation of DOR-induced MAP kinase and M3-muscarinic receptor-induced cAMP accumulation but had no effect on α₂-adrenergic receptor or bradykinin receptor signaling (181). These results once again exemplify how a GPCR can modulate RGS protein levels to cross-regulate other GPCRs. As another example, constitutively active Gα_o induced RGS20 ubiquitination and downregulated RGS20 expression through proteasomal degradation (121). In the same study, activation of the Gα_o/i-coupled serotonin receptor led to RGS20 degradation, which resulted in loss of inhibition of G_i signaling and was reversed upon addition of the proteasome inhibitor lactacystin.

Lastly, there is one report (134) on sumoylation of RGS proteins, RGSZ1 and RGSZ2, which are conjugated to SUMO1, 2, and 3 in synaptosomal membranes of the periaqueductal gray matter. The sumoylated Rz proteins can coprecipitate with Gα_z and μ-opioid receptors, thus serving as a scaffold for this complex. Administration of morphine before membrane extraction increased the Gα_z/i and RGSZ2 interactions, which peaked thirty min posttreatment, corresponding to the maximal point of morphine-induced analgesia (134).

Overall, the literature demonstrates that posttranslational modifications can modulate RGS activity via multiple modes of regulation such as intracellular localization, protein-protein interactions, and protein stability. The data also indicate complexity within these mechanisms; for example, the effect of phosphorylation or palmitoylation often depends on the specific residue that is modified.

Conclusions

This review has summarized the current information on the regulation of RGS proteins through their expression, intracellular localization, and covalent modifications. Overall, the literature has demonstrated that regulation of RGS proteins occurs on multiple levels and offers a complexity that allows for plastic control of G-protein function. However, it is anticipated that more studies will be performed in the following directions, as discussed below.

While multiple studies report the regulated expression of RGS mRNA, it is important to assess RGS expression at the protein level as well (which a few studies have convincingly demonstrated, see Table 1). One study exemplifies the importance of this notion by showing that cAMP signaling has opposite effects on RGS13 mRNA and protein expression (194). However, the availability of specific antibodies against some RGS proteins remains a significant obstacle. Furthermore, while many studies demonstrate stimuli-regulated RGS expression and concomitant changes in G-protein signaling, the specific requirement for individual RGS proteins was largely not examined. Additionally, microRNAs (miR) represent another mechanism of regulated expression that was not discussed in this review, as at this time, only RGS2 (87, 111) and RGS17 (162) genes have been reported to be regulated by miRs.

Many investigations have explored functions through overexpression of RGS genes; however, it is essential to examine if the conclusions translate to the endogenous RGS proteins. While the overexpression of several RGS proteins have proven valuable in characterizing their biochemical properties, this approach cannot ascertain whether RGS proteins function redundantly. Additionally, ectopic expression may prevent accurate interpretation of protein function, for instance, by potentially negating regulatory mechanisms. The function of endogenous RGS proteins can be addressed through studies of RGS knockout mice, which provide the additional advantage of examining in vivo function. Importantly, heterozygotes should be also examined to evaluate the effect of partial downregulation of the RGS protein, which would support or refute the functional significance of partial RGS downregulation observed in a number of cell culture studies. However, given that several RGS proteins may be regulated in a similar manner in the same cell type, and given the functional redundancy of some RGS proteins, the double/triple knockout mice would have to be generated. RNA interference-mediated knockdown can serve as an alternative method to examine the endogenous RGS proteins, and it offers the advantage of circumventing the possibility of compensation mechanisms (i.e., possible upregulation of other RGS family members) in knockout mice. However, the potential off-target effects of siRNA were not considered in many of the discussed studies and should be controlled for by using separate siRNAs against different regions of the gene and by rescue experiments using siRNA-resistant RGS mutants. When possible, highly stringent studies should utilize complementary techniques for understanding the RGS regulation and function. In addition, recent development of pharmacological inhibitors of RGS proteins may provide an alternative useful method to study the role of endogenous RGS proteins (154, 155). However, while the in vitro efficacy and specificity of some molecules are promising, their in situ and in vivo efficiency requires validation.

Finally, we anticipate more studies on RGS regulation and function in animal models and humans as they relate to the pathogenesis of disease. Emerging data indicate that RGS proteins are subjected to regulated expression in various diseases, including cardiovascular pathologies and neurological disorders, and are becoming increasingly appreciated in several types of cancers (155, 199). We expect that RGS dysregulation will be demonstrated in other diseases, and the specific contribution of RGS proteins to the pathogenesis of disease will be investigated, which may point to a given RGS as a potential therapeutic target for a disease treatment (154, 155).

GRANTS

This study was supported by National Institutes of Health Awards R01-GM-85058 (to N. O. Dulin), T32-HL-007237 (to J. Kach), and T32-HD-009007 (to N. Sethakorn); and American Heart Association Fellowships 10PRE4190120 (to J. Kach) and 10PRE2630163 (to N. Sethakorn).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: J.K., N.S., and N.O.D. drafted manuscript; J.K., N.S., and N.O.D. edited and revised manuscript; J.K., N.S., and N.O.D. approved final version of manuscript.