RGS5 Attenuates Baseline Activity of ERK1/2 and Promotes Growth Arrest of Vascular Smooth Muscle Cells

The regulator of G-protein signaling 5 (RGS5) acts as an inhibitor of Gαq/11 and Gαi/o activity in vascular smooth muscle cells (VSMCs), which regulate arterial tone and blood pressure. While RGS5 has been described as a crucial determinant regulating the VSMC responses during various vascular remodeling processes, its regulatory features in resting VSMCs and its impact on their phenotype are still under debate and were subject of this study. While Rgs5 shows a variable expression in mouse arteries, neither global nor SMC-specific genetic ablation of Rgs5 affected the baseline blood pressure yet elevated the phosphorylation level of the MAP kinase ERK1/2. Comparable results were obtained with 3D cultured resting VSMCs. In contrast, overexpression of RGS5 in 2D-cultured proliferating VSMCs promoted their resting state as evidenced by microarray-based expression profiling and attenuated the activity of Akt- and MAP kinase-related signaling cascades. Moreover, RGS5 overexpression attenuated ERK1/2 phosphorylation, VSMC proliferation, and migration, which was mimicked by selectively inhibiting Gαi/o but not Gαq/11 activity. Collectively, the heterogeneous expression of Rgs5 suggests arterial blood vessel type-specific functions in mouse VSMCs. This comprises inhibition of acute agonist-induced Gαq/11/calcium release as well as the support of a resting VSMC phenotype with low ERK1/2 activity by suppressing the activity of Gαi/o.


Introduction
Vascular smooth muscle cells (VSMCs) form the contractile layer of the blood vessel wall and specifically respond to extra-and intracellular stimuli by increasing their tone to adapt blood flow and pressure. Stimuli by humoral agonists are usually sensed by corresponding receptors, which initiate signaling cascades to activate the contractile apparatus of the VSMCs. Heterotrimeric G-proteins-consisting of α subunits and the associated βγ-subunits-are important regulators of such signaling cascades determining vascular tone [1]. Their activation is usually initiated via binding of agonists to corresponding G-protein coupled receptors (GPCRs), which are capable to interact with one or several types of G-proteins including Gα q/11 , Gα 12/13 , Gα i/o , and Gα s . In their inactive state, Gα-subunits are bound to guanosine diphosphate (GDP). Upon interaction with a ligandstimulated GPCR, GDP becomes replaced by guanosine triphosphate (GTP) allowing the Gα subunits to interact and activate their specific target effector proteins [2]. For instance, while stimulation of Gα q/11 triggers the release of calcium from intracellular stores to promote acute contraction of VSMCs, Gα s -mediated signals have the opposite functional effect and may be counteracted by activation of Gα i/o .
In addition to the agonist-induced VSMC contraction, G-protein signaling governs the activation of mitogen activated protein (MAP) kinases [3], which regulate several critical cellular functions, such as cell proliferation, differentiation, and apoptosis [4]. Upon activation, these kinases translocate from the cytoplasm to the nucleus and via phosphorylation control the activities of various transcription factors, subsequently mediating the abovementioned cellular processes [5,6]. G-protein and MAP kinase signaling are connected via multiple pathways, including the activation of Ras via βγ-subunits [7], the activation of C-Raf by the Gα q -PKC axis and the enhancement of C-Raf activity by suppression of two, PKA and Rap1-dependent, inhibitory pathways via Gα i [8]. Considering the relevance of these interconnected pathways for the control of cellular functions, the regulatory impact of G-protein-mediated signaling on acute and long-term responses of VSMCs to environmental stimuli is thus subject to further investigation.
Although primarily triggered by agonists or environmental stimuli, the overall activity of the Gα-subunits and associated signaling events is controlled by regulators of G-protein signaling (RGS) [9,10]. Proteins of this family comprise at least 25 members with a characteristic GTPase-activating protein (GAP) domain that accelerates the hydrolysis of GTP bound to Gα subunits, promoting their inactive GDP-bound state and terminating the corresponding signal transduction [11,12]. As such, RGS proteins may shift the outcome of any G-protein-mediated signaling [13,14] and modulate acute responses of VSMCs to environmental stimuli as well as the baseline activity of individual signaling pathways. Based on previous findings, genetic ablation of Rgs2 in mice prolonged preferentially Gα q/11 -mediated signaling and caused hypertension [15] indicating its general relevance for the basal tone of VSMCs. In contrast, loss of RGS5-a potent inhibitor of Gα q/11 and Gα i/o activity-resulted in hypo- [16], normo- [17], and hypertensive [18,19] mice depending on their genetic background and the applied methodology. Moreover, proliferation of VSMCs is controlled by RGS5 during neointima formation [20] and our own studies revealed the ability of RGS5 to stimulate RhoA activity in VSMCs in the context of arterial hypertension [21].
As can be deduced from the aforementioned findings, the regulatory influence of RGS5 on acute and baseline VSMC signaling, as well as its G-protein-specific regulatory features, are still under debate. In this context, we assumed that the reported partially contradicting functional features of RGS5 might be explained by context-specific inhibitory functions, which may promote diverging effects. Therefore, this study was intended to delineate longterm influence of RGS5 on different signaling pathways as well as on the functional status of resting and proliferating VSMCs by applying loss-and gain-of-function approaches. The observed effects were compared with those induced by selectively inhibiting the activity of Gα q/11 and Gα i/o subunits to evaluate their individual contribution to the regulatory features of RGS5.

Telemetric Blood Pressure Measurement
Blood pressure measurements were performed in global-or SMC-specific Rgs5deficient mice by implanting telemetric devices (PA-C10; Data Sciences International, St. Paul, MN, USA) as previously described [22]. In brief, through the left common carotid artery, catheter tips were advanced into the aortic arch of 12-week old mice. A signal transducer unit was placed in a subcutaneous pocket on the right ventrolateral side of the mouse. One week (recovery time) after surgery systolic/diastolic blood pressure measurements were recorded every 30 min for 5 min using Dataquest A.R.T. software 4.0.

Cell Culture
The aorta of floxed Rgs5 (Rgs5 fl/fl ) mice were used to isolate murine aortic SMCs (aoSMC). In brief, after careful removal of the surrounding adventitial tissue, the aorta was washed twice in Dulbecco's PBS (without calcium and magnesium), cut in 1-mm-sized rings, and digested overnight with 1% collagenase (#C5138, Merck GmbH, Darmstadt, Germany). Suspended cells were seeded on culture plates, routinely checked for marker expression (e.g., αSMA, SMMHC), and used only until passage 5 throughout the experiments. The procedure for the human umbilical artery smooth muscle cell (HUASMC) isolation were carried out with the approval from the Local Ethical Committee (date: 12 April 2013/ID: S-191/2013, document number: 336/2005, Heidelberg, Germany) and conformed to the principles outlined in the Declaration of Helsinki (1997). HUASMCs were either isolated from human umbilical cord arteries or purchased from ProVitro (Berlin, Germany) and were cultured up to passage 5. All VSMCs were cultured in low-glucose DMEM (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 15% fetal calf serum (Biochrom, Cambridge, UK).

Adenoviral Transfection
HUASMCs were transduced using adenoviral vectors (MOI 200, kindly provided by Prof. Dr. Thomas Wieland, Mannheim, Germany), which have been described earlier [17] to overexpress only GFP as a control (Ad-GFP) or His-tagged RGS5 and GFP (Ad-RGS5) both under a CMV promoter. Cells were transduced at 70% confluency in medium supplemented with 2 µg/mL polybrene ® (sc-134220, Santa Cruz Biotechnology, Inc., Heidelberg, Germany) to improve transfection efficiency. Cells were washed after 18 h to remove the viruses.

Generation of 3D Spheroids
3D cellular-spheroids were generated using hanging drops methodology. Briefly, cells were detached with trypsin, centrifuged at 300× g for 5 min, resuspended and counted in a Neubauer chamber. A total of 3000 (HUASMC) or 500 cells (aoSMC) were mixed per 25 µL medium with 0.24% w/v methyl cellulose (#M0650, Merck, Darmstadt, Germany) and 15% FCS and spotted dropwise onto flat petri dishes. Petri dishes were turned upside down and incubated for 48 h to generate hanging drops. Petri dishes were washed with 10 mL of DPBS (#14040091, Thermo-Fisher, Waltham, MA, USA) to harvest the 3D spheroids, which were centrifuged at 300× g for 5 min and then processed for mRNA or protein analyses.

3D Migration Assay
The 3D spheroids (500 cells/spheroids) were generated by hanging drop method for 24 h and subsequently embedded into collagen matrices. A total of 4.5 mL acidic collagen extract of rat tails (5 mg/mL, 4 • C) was mixed with 500 µL of 10× M199 (#M0605, Sigma-Aldrich), and titrated with 0.2 M NaOH (approximately 500 µL) to neutralize the stock solution. This stock solution was then mixed with equal volume of DMEM (containing 30% FCS, 1.2% w/v methyl cellulose) and the 3D spheroids (about 75 spheroids/mL). This mixture (1 mL/well) was transferred into pre-warmed 24-well plates and incubated at standard conditions for 30 min. Test substances dissolved in 0.1 mL DMEM/ 15% FCS were pipetted on top of each well. After 24 h, the cumulative length of the sprouts originating from one spheroid was determined using an Olympus CKX 40 microscope (Olympus CACh 10x/0.2 NA dry objective lens, Olympus U-LS03-3 camera) and the imaging software CellD (v.3.4; Olympus, Hamburg, Germany). For each experimental group, 10 spheroids were analyzed and an average cumulative sprout length per spheroid was determined.

Microarray-Based Transcriptome Analysis
HUASMCs were transduced with adenoviruses encoding GFP or RGS5 as described. Total RNA from these cells was isolated using the RNeasy Mini Kit (#74106, Qiagen, Hilden, Germany). Gene expression profiling was performed using arrays of human HuGene-2.0-st-type (#902112, Thermo Fisher Scientific, Waltham, MA, USA). Using Affymetrix standard labelling protocol utilizing the GeneChip WT Plus Reagent Kit and the GeneChip Hybridization, Wash and Stain Kit (Affymetrix, Santa Clara, CA, USA), biotinylated antisense cDNA was prepared. Hybridization on the chip was carried out in a GeneChip Hybridization oven 640 (Affymetrix), followed by dyeing in the GeneChip Fluidics Station 450 (Affymetrix) and thereafter scanned with a GeneChip Scanner 3000 (AffymetrixA custom Chip Design File (v.22, CDF)) with Entrez-based gene definitions was used to annotate the arrays [23]. The raw fluorescence intensity values were normalized applying quantile normalization and RMA background correction. One-way ANOVA was performed to identify differentially expressed genes using a commercial software package JMP10 Genomics v.6 from SAS (SAS Institute, Cary, NC, USA). A false positive rate of a = 0.05 with false-discovery rate correction was taken as the level of significance. Gene Set Enrichment Analysis (GSEA) was performed to determine whether defined lists (or sets) of genes exhibit a statistically significant bias in their distribution within a ranked gene list. The statistical procedure described was performed using the software GSEA. Pathways belonging to various cell functions, such as cell cycle or apoptosis, were obtained from public external databases (KEGG, http://www.genome.jp/kegg, accessed on August 2019).

Quantitative Real Time RT-PCR (qPCR) Analysis
Murine arteries were isolated and frozen in QIAzol Lysis reagent (#79306, Qiagen, Hilden, Germany). The tissue was disrupted using a tissue lyser (Retsch, MM301, 30 Hz, 1.5 min) and 5 mm stainless steel beads. For RNA isolation from cells, RLT Buffer (#79216, Qiagen) containing 1% β-Mercaptoethanol was used to lyse the cells. Total RNA was isolated by solid-phase extraction using the RNeasy ® Micro Kit (#74004, Qiagen) by following the manufacturer's instructions. Subsequently, cDNA was synthesized using the Sensiscript Reverse Transcription Kit (#205213, Qiagen). Quantitative real-time RT-PCR for the target sequences (primer sequences are listed in Table 1) was performed in the Rotor-Gene Q (Qiagen) using the LightCyclerR 480 SYBR Green I Master Mix (#04707516001, Roche, Mannheim, Germany). Fluorescence was monitored (excitation at 470 nm and emission at 530 nm) at the end of the annealing phase. Threshold cycle (Ct) was set within the exponential phase of the PCR. Quantification of the PCR product was done by using the ∆∆Ct method. Amplification of the 60S ribosomal protein L32 (RPL32) cDNA served as an internal reference. Table 1. PCR primer sequences.

Gene Name
Sequence Annealing Temperature
Capillary electrophoresis (Figures 2, 3 and 6e, Supplementary Figure S2) was used for separating and automated detection of proteins in RIPA lysates employing the WES™ system (ProteinSimple, San Jose, CA, USA). Reagents and samples were prepared according to manufacturer's instructions. After dilution of samples to a final concentration of 1 mg/mL, protein separation, and detection steps were carried out with the following settings: separation voltage, 375 V; separation time, 30 min; Antibody diluent time, 5 min; primary antibody time, 90 min; and secondary antibody time, 30 min. For capillary electrophoresis, following dilutions of primary antibodies were used: VCP, 1:50; RGS5, 1:25; ERK1/2, 1:50; and pERK1/2, 1:50. Signal intensities of bands were automatically analyzed using the Compass SW software (v3.1.7; ProteinSimple) and, protein levels were reported relative to appropriate loading controls.

Human Phosphokinase Array
Proteome Profiler Human Phospho-Kinase Array Kit (#ARY003B, R&D Systems ® , Minneapolis, MN, USA) was utilized to simultaneously determine the relative phosphorylation levels of several kinases in the cell lysates by following manufacturer's instructions. Briefly, cell lysates were incubated with the membranes where specific phospho-specific primary antibodies have been spotted. After washing steps, membranes were incubated with biotinylated detection antibodies followed by Streptavidin-HRP. Chemiluminescent detection was carried out using the Image Quant™ TM LAS 4000 mini (GE Healthcare, Waukesha, WI, USA). Grey intensities were quantified using the ImageJ Software (NIH, USA, Version 1.49a).

Measurement of Intracellular Calcium Mobilization
Cells or spheroids were incubated with 2.5 µM Rhodamine-4 AM (AAT BioQuest, Sunnyvale, CA, USA) dissolved in HBSS Buffer with Ca 2+ /Mg 2+ (PAA, Pasching, Austria) containing 20 mM HEPES and 0.001% detergent (Pluronic F12, AAT BioQuest) for 30 min at 37 • C and followed by another 30 min at RT in the dark. Angiotensin II (#A6402, Sigma) and Norepinephrine (#A0937, Sigma)-evoked mobilization of intracellular calcium was measured using the fmax Fluorimeter (Molecular Devices, Biberach, Germany). Relative fluorescence units were measured every 5 s for a period of 5 min and were normalized to non-stimulated controls.
Fluorescence intensity of spheroids was determined by live cell fluorescence imaging. To this end, spheroids were transferred to 96 well plates, incubated at 37 • C, 5% CO 2 in a Tokai Hit INU incubation unit, and stimulated with the indicated agonists. The fluorescence intensity of the spheroid was recorded every second by a Hamamatsu ORCA-R2 C10600-10B camera and quantified by utilizing the cellSens dimension software (v.1.12; Olympus).

RhoA Activity Assay
GTP-bound fraction of RhoA was measured, using the G-LISA RhoA Activation Assay Biochem Kit (Cytoskeleton, Inc., TebuBio, Offenbach, Germany). HUASMCs were cultured in 6-well plates and, subsequently, transfected with either control siRNA or RGS5-targeting siRNA to deplete RGS5 Cells, were lysed, and the protein concentration was determined and adjusted for further analyses according to the manufacturer's instructions. GTP-bound RhoA was normalized to the total RhoA content of the samples, which was assessed by Western blot analysis.

Immunofluorescence-Based Detection of Ki67
HUASMCs were washed with HBSS, and fixed with 100% methanol (−20 • C) for 15 min and dried. Fixed cells were blocked for 1 h in 0.25% casein and incubated with the anti-Ki67 antibody (1:2500) at 4 • C overnight. After three washing steps, cells were incubated with the secondary antibody for 1 h at RT. After staining of the nuclei with DAPI, the fluorescence was recorded using Olympus IX83-System with cellSens Software (v. 1.12). The percentage of Ki67-positive nuclei was quantified by utilizing the ImageJ Software (v 1.49).

Statistical Analysis
All results are reported as means ± SD and statistically analyzed by utilizing Graph-Pad Prism (Version 9.1). Outliers were identified by application of the Grubbs' test with α set to 0.05. Differences among normally distributed values of two matched experimental groups were analyzed by unpaired Student's t-test or one sample t-test if appropriate. p < 0.05 was defined as statistically significant. Differences of one parameter between normally distributed values of three or more experimental groups were analyzed by oneway ANOVA followed by Šídák's multiple comparisons test for selected pairs of groups or a Tukey-Kramer test for multiple comparisons. p < 0.05 was considered statistically significant (* p < 0.05, ** p < 0.01, *** p < 0.001).

Regulatory Features of RGS5
RGS5 inhibits the activity of both Gα q/11 and Gα i/o -dependent signaling pathways and, thus, may decrease the tone of VSMCs by simultaneously attenuating Gα q/11 -PLCinduced IP3/calcium-and DAG/PKC-mediated contraction, and by limiting the inhibition of Gα i/o -adenylyl cyclase-cAMP-PKA-mediated relaxation. Moreover, the inhibition of these signaling pathways may also affect the activity of RhoA-and mitogen activated (MAP) kinases, which are connected to the G-protein activity via βγ-subunits, PKA and PKC [8]. While these regulatory features illustrate the fundamental capacity of RGS5 for controlling VSMC functions, the actual inhibitory impact may be dependent on its contextand cell-specific expression.

Heterogeneous Artery Type-Specific Expression of Rgs5 in Mice
Considering the diverse developmental origin of different VSMC populations associated with varying levels of Rgs5 expression [24], we assumed artery type-specific differences in the Rgs5 expression of mouse arteries. Corresponding analyses of different mouse arteries included small arteries, such as third order branches of the mesenteric artery and large conduit arteries such as the aorta. These exhibited marked differences in the levels of Rgs5 expression (Figure 1a), which appeared to correlate, at least in part, with the detectable protein levels (Figure 1b). In contrast, the expression levels of Rgs2 and Rgs16 were much lower and homogeneous as compared to Rgs5 (Figure 1a).

Ablation of Rgs5 Does Not Alter the Systolic and Diastolic Blood Pressure Values but Stimulates ERK1/2 Phosphorylation
Based on these findings, we investigated whether loss of Rgs5 affects the arterial baseline tone in mice. We indirectly assessed this parameter by performing a continuous telemetry-based analysis of the blood pressure in wild type (WT) and Rgs5-deficient (Rgs5 −/− ) mice. No differences were detected for the averaged systolic and diastolic blood pressure values (Figure 2a). Similar results were obtained with mice upon SMC-specific ablation of Rgs5 (Figure 2b, Supplementary Figures S1 and S2a), indicating that RGS5mediated effects in other cell types (e.g., endothelial cells) may exert a negligible impact on the general vascular tone.
Considering that knockout of Rgs2-another inhibitor of Gα q/11 and Gα i/o -signalingcauses an increase in baseline blood pressure [15], these results suggest an artery-type specific functional role of RGS5. We assumed that arteries showing low expression of Rgs5 such as branches of the mesenteric artery were less affected by its knockout than arteries with robust Rgs5 expression such as large conduit arteries. Consequently, we selected the caudal artery and aorta to investigate whether loss of Rgs5 may influence cellular signaling cascades. We focused on the MAP kinases ERK1/2 as their activity is known to be modulated by G-proteins [8] as well as RGS5 [25] and found an increased level of ERK1/2 phosphorylation after SMC-specific knockout of Rgs5 (Figure 2c).

High Levels of RGS5 Are Maintained in Mouse and Human VSMCs under 3D Culture Conditions In Vitro
For further mechanistic analyses addressing the consequences of Rgs5 deficiency, we required an experimental setup that maintains Rgs5 expression in cultured mouse aortic VSMCs. We applied a 3D spheroid culture technique that allows the formation of multiple SMC layers and promotes a differentiated, contractile and quiescent VSMC phenotype [26]. As compared to proliferating murine VSMCs cultured under 2D conditions, Rgs5 expression was upregulated in 3D spheroids (Figure 3a) while its knockout (Figure 3b) was associated with increased ERK1/2 phosphorylation (Figure 3c). Comparable results were obtained with human VSMCs maintaining high levels of RGS5 in 3D spheroids (Figure 3d,e) and significantly reduced ERK1/2 phosphorylation (Figure 3f) as compared to VSMCs cultured as 2D monolayer.

Overexpression of Rgs5 and Inhibition of Gα i/o Signaling Inhibits ERK1/2 Activity, Proliferation and Migration of Human VSMCs
The aforementioned findings suggested that in resting VSMCs, high levels of RGS5 may functionally contribute to cellular quiescence and differentiation e.g., by attenuating the impact of stimulating agonists and supporting RhoA activity [27] while limiting the activity of MAP kinases. Elevated expression of Rgs5 in active and proliferating VSMCs may thus in turn force their quiescent state. To test this hypothesis, we overexpressed RGS5 in 2D cultured proliferating VSMCs (Supplementary Figure S3) and investigated alterations of their transcriptome by applying whole genome microarray analyses (Supplementary Figure S4). Based on the outcomes of the microarray analyses, RGS5 decreased the level of transcripts associated with DNA replication and cell cycle control (Figure 5a,c) including CCNE2, CDC6, and Ki67 (Figure 5b) and stimulated the expression of genes encoding cytokines or molecular components controlling cell adhesion and interaction with the extracellular matrix (Figure 5a,b,d). In addition, RGS5 also attenuated the expression of genes associated with the metabolism (butanoate, purine, pyrimidine), as well as translation and RNA processing, suggesting a general decline in cellular activity. Among the top upregulated transcripts were KCNK2-encoding a mechanosensitive potassium channel [28], and MIR145-a key mediator of smooth muscle cell differentiation [29]. However, the gene set attributed to "vascular smooth muscle contraction" was not significantly altered (NES: 1.17, adj. p-val. 0.41). Likewise, no significant changes in expression were observed for prototypic smooth muscle cell differentiation marker genes such as MYH11, ACTA2, CNN1, or TAGLN.  On the signaling level, Rgs5 overexpression decreased the phosphorylation of several protein kinases, such as Akt (Figure 6a) and the MAP kinases ERK1/2 and JNK (Figure 6b), which was determined by quantitating their phosphorylation of activating phosphorylation sites. Correspondingly, RGS5 elevated the phosphorylation level of CREB and c-Juntargets of Akt and MAP kinases, respectively. We next investigated which G-protein-dependent signaling pathway may mediate the observed responses of VSMCs by selective inhibition of either Gα i/o or Gα q/11 activity. Pertussis toxin (PTX)-mediated inhibition of Gα i/o , but not YM254890-based interference with Gα q/11 activity (Supplementary Figure S5) decreased ERK1/2 phosphorylation (Figure 6c-e). In addition, proliferation and migration of VSMCs were analyzed as functional readouts representing the activated VSMC phenotype. Immunofluorescence-based detection of the proliferation marker Ki67 in the nuclei of VSMCs indicated an inhibitory influence of RGS5 on the cell cycle ( Figure 7a). Furthermore, RGS5 mitigated the outgrowth of VSMCs from VSMC-spheroids into a 3D collagen matrix (Figure 7b). Again, both inhibitory effects were mimicked by interfering with the activity of Gα i/o but not Gα q/11 (Figure 7c,d) or by stimulating adenylyl cyclase through forskolin (Supplementary Figure S6).

Discussion
The phenotype and function of VSMCs is defined by a plethora of developmental and environmental determinants (e.g., blood pressure). As such, arterial VSMCs of the adult vasculature represent a phenotypically diverse cell population with organ-specific functional features. The diversity of the VSMC phenotype was recently highlighted by single cell RNA-sequencing analyses, which revealed that the gene expression and protein pattern of GPCRs is highly diverse even within a population of VSMCs from the same type of artery [30]. For instance, heterogeneous expression was observed for most GPCR transcripts in both large caliber conductance and small caliber resistance arteries, while some GPCRs, such as Ednrb (endothelin receptor type-B) appeared preferentially expressed in the latter. Given the complexity of the resulting signaling scenario for each VSMC subpopulation, the relevance of RGS proteins for balancing the activity of individual signaling pathways becomes even more evident. To this end, our results indicate an artery subtype-specific expression of RGS5 thereby verifying earlier reports suggesting that the developmental origin of VSMCs determines the RGS5 expression level [24]. In line with our observations, in silico analyses utilizing a database for tissue-specific gene expression [31] revealed highly heterogeneous expression levels of RGS5 in human arteries (Supplementary Figure S7). Moreover, single cell RNAseq-based data [32] from mice also suggests diverse expression levels of Rgs5 even within VSMC populations from one organ. For instance, Rgs5 is strongly expressed in venous and arteriolar VSMCs of the brain but barely detectable in arterial VSMCs and endothelial cell (EC) types (Supplementary Figure S8). Such findings also imply a minor relevance of RGS5 for EC signaling in mice, at least under baseline conditions. Correspondingly, our study for the first time shows comparable arterial baseline blood pressure values in mice with SMC-specific and systemic knockout of Rgs5, suggesting that additional loss of endothelial RGS5 in the latter mouse line does not affect blood pressure regulation. Moreover, our results also imply that genetic ablation of Rgs5 evokes an artery subtype-specific effect rather than a general lapse of systemic blood pressure regulation.
With respect to mechanisms controlling the expression of Rgs5 in mice, several molecular and environmental determinants have been identified. On the epigenetic level, methylation of its promoter may control Rgs5 expression in a developmental origin-specific manner [24]. Arterial hypertension appears to downregulate Rgs5 expression in angiotensin II-treated mice [18] while the opposite was observed in deoxycorticosterone acetate (DOCA)-treated hypertensive mice [21] indicating a context-dependent regulation. On the cellular level, Rgs5 expression was increased in VSMCs or arteries exposed to nitric oxide or biomechanical stretch [17,21] and repressed in VSMCs under mitogenic culture conditions [20] or after treatment with platelet-derived growth factor [33]. Here, we show that the low expression level of RGS5 in proliferating VSMCs is robustly increased under 3D culture conditions supporting the resting VSMC phenotype [26]. In this context, RGS5 appears to act as a determinant of cellular quiescence as it attenuates Gα q/11 and Gα i/oprotein activity, and significantly alters the activity of downstream signaling cascades. For instance, RGS5 promotes the baseline activity of RhoA confirming earlier observations and its link with Gα i/o and Gα q/11 -mediated signaling events [21]. Functionally, RhoA activity stimulates the expression of prototypic SMC markers, such as αSMA and SM22 [27]-a hallmark of a contractile and quiescent VSMC phenotype that predominates under the chosen 3D culture conditions [26]. Moreover, RhoA may also contribute to transforming growth factor-β-dependent Smad signaling, which supports VSMC differentiation [34,35] and is critical for preserving the growth arrest in 3D VSMC spheroids [26]. Inhibitory features of RGS5 described within this study apply to both, the acute agonist-induced stimulation of Gα q/11 -dependent calcium release from intracellular stores and the baseline activity of the MAP kinase ERK1/2. The former observation relies on the well-described termination of Gα q/11 -mediated calcium mobilization [13,17], which may also affect the recently described Gα q/11 -restricted Gα i -Gβγ-PLCβ-mediated calcium release [36]. Likewise, a principal link of ERK1/2 phosphorylation and G-protein signaling has been repeatedly reported for different experimental conditions and cell types [13,15,31,32]. Several G-protein-associated signaling sources may be involved such as the Gβγ-Ras axis [3], the Gα q/11 -PKC axis [37,38] and, though in an inverse fashion, also the Gα i/o -PKA axis considering that PKA inhibits Gα q/11 -mediated signaling cascades [39][40][41].
Interestingly, the inhibitory effects of RGS5 on the acute agonist-induced calcium release is abolished upon selective inhibition of Gα q/11 while baseline MAP kinase activity is only blocked by selective inhibition of Gα i/o but not Gα q/11 . This implies as yet unrecognized agonist-independent G-protein-associated MAP kinase activity in mouse and human VSMCs that is mainly balanced via Gα i/o -dependent signaling pathways. Thus, suppression of Gα i/o activity through RGS5 would enhance the adenylyl cyclase-cAMP-PKA signal strength with the capacity to limit MAP kinase activation, as was evidenced by the forskolin-induced inhibition of ERK1/2 phosphorylation. Additionally, RGS5 would also abrogate Gα q/11 -PKC-as well as Gβγ-dependent MAP kinase activation since termination of Gα activity is known to preclude the dissociation of the corresponding Gβγ-subunits. However, the overall impact on ERK1/2 phosphorylation brought about by functionally inhibited Gα q/11 was minor. This may also provide a possible explanation why selective inhibition of Gα q/11 potently interferes with the directly linked agonist-induced calcium release but not with baseline MAP kinase activity as an indirectly connected downstream signaling event that is controlled by many additional signaling cascades. In fact, restricting Gα i/o activity has a much greater relevance in this context.
On the functional level, RGS5 has been reported to control many aspects of the VSMC phenotype. For instance, its overexpression inhibits VSMC proliferation during neointima formation in mice [20]. Comparable results were obtained in this study whereby overexpression of RGS5 inhibited proliferation and migration of cultured human VSMCs. However, phenotype changes of VSMCs evoked by biomechanical stretch or wall stress as it occurs during hypertension-induced structural remodeling of the arterial wall were impaired in mice upon knockdown of Rgs5 [21]. It may thus be tempting to speculate that chronically amplified MAP kinase activity caused by the loss of RGS5, as we have observed it in vitro and in vivo, may indeed promote the activated and synthetic phenotype of VSMCs. In the long run, this maladjustment may support thickening and stiffening of the media. Considering the artery type-specific expression of RGS5, the relevance of such remodeling processes for individual sites of the arterial system may nevertheless be highly diverse.

Conclusions
Collectively, RGS5 acts as a context-sensitive artery type-specific regulator that adjusts acute and long-term G-protein-mediated signaling in VSMCs. The inhibition of Gα i/o and Gα q/11 -dependent signaling caused by its expression supports the resting phenotype of VSMCs by silencing MAP kinase activity, likely by controlling Gα i/o -dependent signaling pathways. In contrast, SMC-specific loss of RGS5 stimulates baseline MAP kinase activity of arteries without affecting the systemic arterial blood pressure.